Continuing the development of the public service logic: a study of value co-destruction in public services

ABSTRACT

This paper reports on a study of value co-destruction in public services, i.e. diminishment of value by interaction between providers, users, and other actors. The goal is to contribute to the public service logic (PSL) that suggest a shift from linear co-production to dynamic value co-creation. However, PSL has devoted scant attention to value co-destruction. The paper contributes by identifying four dimensions representing causes of value co-destruction in public services. The paper also shows how value may be co-destroyed in the interaction between several types of actors, thus advancing a service ecosystems perspective for understanding value co-destruction.

KEYWORDS:

• Public service logic
• value co-destruction
• value co-creation
• service ecosystems

Introduction

Public management scholars have outlined the public service logic (PSL) in order to understand the co-production and management of public service organizations (PSOs) (e.g. Alford 2014, 2016; Hardyman, Daunt, and Kitchener 2015; Osborne 2010, 2018; Osborne, Radnor, and Strokosch 2016; Radnor et al. 2014). The departure point for this stream of research is the argument that current public management theory is not fit for purpose because it is rooted in management theory deriving from goods-producing firms ignoring the fact that the public sector is devoted to facilitating services for citizens. Inspired by developments in service research (e.g. Grönroos 2011, 2019; Grönroos and Voima 2013; Vargo and Akaka 2012; Vargo and Lusch 2004, 2008, 2016), the alternative service-centred PSL has been articulated.

The PSL links co-production directly to the co-creation of value focusing on how PSOs, citizens, and other relevant public actors integrate resources to co-create value (Osborne 2018; Osborne, Radnor, and Strokosch 2016). Hence, PSL emphasizes a process that increases users’ wellbeing (Grönroos and Voima 2013), and ‘assumes an interactive and dynamic relationship where value is created at the nexus of interaction’ (Osborne 2018, 225). While PSL is a much welcomed alternative to NPM, its treatment of value has also been problematized to some extent by public management scholars. It has been pointed out that the service research informing the PSL has mainly been developed with the private sector in mind, thus focusing on users’ individual and private value. Value in the public sector, it has been argued, needs to be addressed in a broader perspective that includes not only private value but also group value and public value (Alford 2014; Meynhardt 2009). While private value benefits users and is consumed individually by users, public value benefit the society at large and is consumed or received collectively by the citizenry (Alford and Hughes 2008; Moore 1995). Occupying a middle position, group value concerns the community level such as value for a housing cooperative or a parent association at a school. Examples of public value include care of the environment, securing people’s rights and justice, equal treatment, equal access to services and the upholding of democratic principles (Alford and Hughes 2008; Jørgensen and Bozeman 2007). Alford (2016, 682) emphasizes the fact that ‘public and individual values are usually produced together but discerned through radically different processes, and consumed by different actors’. When developing PSL further through empirical studies (e.g. Eriksson 2019; Hardyman, Daunt, and Kitchener 2015), it is important to take the prerequisites of the public sector including the distinction between private and public value, into account (Alford 2009).

In this paper, we contribute to the discourse on the PSL by attending to the dark side of value. In particular, we focus on value co-destruction, a notion that has been receiving increasing attention from service researchers (Echeverri and Skålén 2011; Plé and Chumpitaz Cáceres 2010; Smith 2013) but which has only received scant attention from public management and PSL scholars (for an exception, see Williams, Kang, and Johnson (2016)). The notion of value co-destruction captures the diminishment of value for one or more actors that are involved in direct interactions with each other (Echeverri and Skålén 2011; Smith 2013). Hence, studies of value co-destruction and PSL share an emphasis on individual users and private value, which is therefore also our focus in this paper. As with value co-creation research, studies of value co-destruction focus on the resource integration process that actors linked via direct interactions, engage in. However, studies of value co-destruction focus on the misuse of resources and the resulting decline in wellbeing rather than on increased wellbeing and positive outcomes from resource integration (Plé and Chumpitaz Cáceres 2010). In neglecting value co-destruction, the PSL is running the risk of losing some of its explanatory power due to only dealing with the positives. Our goal, in writing this paper, is to address this lacuna. In particular, we aim to understand how value is co-destroyed in public services due to focusing both on its causes and on identifying which actors may be responsible for causing it.

In order to address this aim, we empirically studied case management at two major Swedish government agencies – the Social Insurance Agency and the Tax Agency – focusing on the direct interaction between these agencies and their users. However, in studying this dyadic interaction, we also learnt that other actors, e.g. the Swedish Public Employment Service, medical services, and employers, played a major role in case management. Therefore, the paper follows the suggestion of Plé and Chumpitaz Cáceres (2010) to address value co-destruction from a service ecosystem perspective, which entails focusing on the resource integration of multiple linked actors (Mars, Bronstein, and Lusch 2012). However, since our empirical study is devoted to the interaction between the agencies under study and their users, only the service ecosystem actors appearing in this interaction are included in the study. Neither does our ecosystems approach include how formal and informal rules and norms impact the co-destruction of value (Akaka, Vargo, and Lusch 2013; Edvardsson, Tronvoll, and Gruber 2011; Vargo and Akaka 2012; Vargo and Lusch 2016). Consequently, this paper explores value co-destruction in what can be called a limited service ecosystem.

The paper contributes by incorporating the notion of value co-destruction with PSL, showing how it can be studied empirically. It also contributes by identifying four dimensions representing the causes of value co-destruction in the public services, and by identifying the actor(s) causing it. The paper further contributes by demonstrating the relevance of informing PSL using an ecosystem view.

The rest of the paper is organized as follows: In the next section, we review the literature on PSL, as well as prior research on value co-destruction. Then we present the methods employed to carry out the present research, followed by a presentation and discussion of the findings. We conclude the paper by articulating our theoretical and practical contributions, and by proposing avenues for further research.

The public service logic

The notion of PSL introduces a break with the NPM-inspired school of thought which has dominated public management theory and practice (Alford 2016; Osborne 2010, 2018). NPM suggests that value is produced within PSOs with little or no involvement on the part of customers, meaning that the focus is intra-organizational. PSL, on the other hand, is informed by service research and suggests that value is co-produced, or co-created, by PSOs and users, as well as third parties (Osborne, Radnor, and Nasi 2012).

Initially, PSL focused primarily on co-production. This means that the PSL was preoccupied with the contributions users make to the service practices planned and executed by the PSO, giving the user an inferior role (e.g. Alford 2016; Osborne, Radnor, and Nasi 2012; Osborne, Radnor, and Strokosch 2016; Radnor et al. 2014). Recently, Osborne (2018, 228) argued for a shift away from value co-production towards value co-creation: ‘PSOs do not create value for citizens – they can only make a public service offering. It is how the citizen uses this offering and how it interacts with his/her own life experience that creates value’ (see also Eriksson (2019); Hardyman, Daunt, and Kitchener (2015); Hardyman, Kitchener, and Daunt (2019); Petrescu (2019); Virtanen and Stenvall (2014), for similar arguments). The notion of co-creation focuses on how organizations engage with their users in order to facilitate, support and enable the value creation experience (Prahalad and Ramaswamy 2004). The shift away from co-production towards co-creation paves the way for more active involvement on the part of the user of public services. For the PSL, the implication of focusing on co-creation, according to Osborne (2018, 5), is that it starts out ‘from the service user as its basic unit of analysis and explores how public services, and PSOs, might be designed to facilitate the co-creation of value by service users, not vice versa’. This conceptualization is in line with the argument made by service researchers (Grönroos and Voima 2013) that actors are resource integrators co-creating value. A distinction is made between two integrated types of resources: i.e. operand resources, which are ‘resources on which an operation or act is performed,’ e.g. ‘land, animal life, plant life, minerals and other natural resources’, and operant resources, which are most prominently the ‘skills and knowledge’ that are ‘employed to act on operand resources (and other operant resources)’ (Vargo and Lusch 2004, 2).

It is also increasingly being acknowledged that resource integration takes place in service ecosystems (e.g. Vargo and Akaka 2012; Vargo, Maglio, and Akaka 2008) involving not only organizations and users, but also different types of connected actors that co-create value (Skålén, Aal, and Edvardsson 2015). The notion of the service ecosystem thus shifts our understanding of value creation away from a dyadic perspective towards seeing value in terms of being co-created by multiple linked resource-integrating actors (Vargo and Lusch 2016; Vargo, Wieland, and Akaka 2016). Several scholars have argued that the notion of service ecosystem can fruitfully inform PSL (Eriksson 2019; Osborne 2018). Petrescu (2019), in a recent conceptual paper, has followed their lead and has explicitly integrated the notion with PSL, thus suggesting that value is co-created through the combined efforts of PSOs, employees, users, political parties, and other relevant stakeholders. In another recent study, Eriksson et al. (2019) demonstrate the usefulness of applying a systems perspective to empirical PSL-informed research. However, PSL studies drawing on the notion of the service ecosystem are still scarce.

The value co-creation of actors in service ecosystems may end up in stable resource configurations, referred to as value propositions, which may be enacted by users in order to create value-in-use (Eriksson et al. 2019; Skålén, et al. 2015; Vargo and Lusch 2008). Value propositions are referred to as promises of value since it is the users who subjectively assesses the value-in-use – a provider never embeds value in products or services during the production process (Grönroos and Voima 2013). Consequently, it is the PSO that must be added to the equation as a co-creator of value, not the service user (Osborne 2018, 5). Users create value on the basis of the value propositions offered by the PSO alone, or combined with other value propositions made by actors within the service ecosystem (Eriksson et al. 2019), and by drawing on their own and/or other users’ resources (Skålén et al. 2018; Trischler and Charles 2019).

Grönroos (2011, 2019) and Grönroos and Voima (2013) offer an illuminating micro-level conceptualization of the resource integration process, suggesting that it can be understood in relation to three different spheres: (i) users’ value creation, (ii), the co-creation of value in direct interaction, and (iii) PSOs’ value facilitation. Value creation refers to the users’ integration of resources during the usage process, without the involvement of PSOs, which is captured by the notion of value-in-use. For example, value creation may concern a user filling in a form to receive unemployment benefit with the aim of receiving this benefit. The co-creation of value concerns all the resource-integrating activities taking place in direct interaction between users and PSOs. A concrete example here is a meeting during which a caseworker helps a user to fill in a form in order to receive unemployment benefit properly. Value facilitation refers to all the initiatives that PSOs conduct internally with the aim of assisting users to create value without directly interacting with them. Examples of this include designing an easy-to-understand form enabling users to apply for unemployment benefit. In short, value facilitation concerns all the internal activities conducted with the intention of offering users attractive value propositions.

While the framework offered by Grönroos (2011, 2019) and Grönroos and Voima (2013) is useful in its simplicity, it has been criticized for being too simple in not taking the wider service ecosystem into account (e.g. Eriksson et al. 2019). Furthermore, the framework also presupposes that actors have fixed roles, which is not always the case; a PSO can take on the role of a user creating value on the basis of the value propositions being offered by another PSO while a user can be a value-facilitator giving feedback to a PSO as regards how to improve its value propositions. Through the service ecosystem lens, it becomes apparent how the same actor can have different roles simultaneously, and at different points in value co-creation processes (Petrescu 2019). In addition, Vargo and Lusch (2016) also offer an understanding of the concept of value co-creation that differs from that of Grönroos. Where Grönroos (2011) states that the user can create value independently of the value facilitator, Vargo and Lusch (2016) argue, on the premise that value is always determined by the user, that the user is always a co-creator of value. In this paper, we follow Vargo and Lusch (2008, 2016).

Value co-destruction

Despite some disagreements, prior research is based on the assumption that value co-creation is a harmonious process resulting in positive outcomes for the actors involved (Grönroos and Gummerus 2014; Vafeas, Hughes, and Hilton 2016). However, service interactions are not always harmonious, they can also be characterized by conflicts, a theme that has been downplayed by service scholars (Laamanen and Skålén 2015). To address this neglect, Echeverri and Skålén (2011) suggest the value-neutral concept of interactive value formation, which takes harmonious and conflictual service interactions as well as positive and negative outcomes into account. We follow the lead of Echeverri and Skålén (2011), paving the way for studying value co-destruction.

According to prior research, value co-destruction takes place between actors involved in a mutual relationship that is based on direct interaction. Value co-destruction happens when interacting parties fail to integrate resources in a mutually-beneficial manner, leading to the diminishment of value-in-use for one or more of the interacting parties (e.g. Järvi, Kähkönen, and Torvinen 2018; Plé and Chumpitaz Cáceres 2010; Worthington and Durkin 2012). Based on a study of direct interactions between providers and users, Echeverri and Skålén (2011) found that value is co-created when these parties enact practices congruently, with the incongruent enactment of practices leading to value co-destruction. Plé and Chumpitaz Cáceres (2010), in another pioneering study of value co-destruction, emphasize that it is an interactional process whereby the misuse of one’s own, or other parties’, resources, either intentionally or unintentionally, creates a decline in at least one of the parties’ wellbeing. According to Vafeas, Hughes, and Hilton (2016), value loss can also occur due to resource deficiency. If one actor fails to share information with another in a timely way, this will be regarded as the accidental misuse of resources. However, if sharing is impossible due to the information being unavailable, this will be considered a resource deficiency. Value co-destruction is also suggested to entail different degrees of value loss. This can range between an absolute loss of value, e.g. an element of a value proposition is lost for good, and a diminishment of value for a shorter period of time (Smith 2013; Vafeas, Hughes, and Hilton 2016).

Extant research, in the private sector context, substantiates the fact that users and providers need to act congruently in interactions in order for value co-creation to ensue (e.g. Echeverri and Skålén 2011; Prior and Marcos-Cuevas 2016; Smith 2013; Vafeas, Hughes, and Hilton 2016). A lack of consonance between actors, as regards what to expect, can cause value co-destruction. Smith (2013) identifies how customers experience the loss of resources (e.g. social, material and in terms of energy) based on the organization failing to respond to customer expectations; also understood as a misuse of their resources by organizations. Smith (2013) and Echeverri and Skålén (2011) study value co-destruction from one side of a dyadic relationship, respectively from the perception of the customer and the employee. The study by Vafeas, Hughes, and Hilton (2016) of both actors in a dyadic relationship (with clients and their creative agencies in a business-to-business context) points to how value loss is caused on the basis of client-, agency- and jointly-situated antecedents.

Although the negative aspects of resource integration have become part of the value discourse, empirical studies of how value can be co-destroyed, and the reasons why, remain scarce (Chowdhury, Gruber, and Zolkiewski 2016; Järvi, Kähkönen, and Torvinen 2018). In addition, the bulk of previous research studies also investigate value co-destruction in dyadic relationships. Plé and Chumpitaz Cáceres (2010) adopted a service ecosystem approach and argued that co-destruction concerns a decrease in the wellbeing of one or more service ecosystem actors. They further suggested that the decrease in wellbeing is a result of the misuse of one or several service ecosystems' resources. Misuse stems from the inappropriate integration of resources by the interacting service ecosystems’ actors, and this can be intended or unintended. However, the study of Plé and Chumpitaz Cáceres (2010) is not based on empirical evidence.

In the public sector context, the service ecosystem perspective on value co-destruction is particularly relevant as multiple inter-independent actors are commonly involved in public service provision (Osborne 2006, 384; Petrescu 2019). However, the scant number of previous studies of value co-destruction in the public sector have not been based on a service ecosystem perspective. Indeed, limited attention has been paid to value co-destruction by public management researchers in general, and by PSL researchers in particular. There are some exceptions, e.g. the study by Williams, Kang, and Johnson (2016), which discusses value (co)-contamination during interactions between service providers and users, in particular the contamination of public value. In addition, Järvi, Kähkönen, and Torvinen (2018) study co-destruction in both private and public organizations. They identify the following eight reasons that lead up to value co-destruction; an absence of information, a lack of trust, mistakes, an inability to serve, an inability to change, an absence of clear expectations, customer misbehaviour, and blaming. While this study provides some valuable insights, it does not discriminate between the public and private contexts and thus does not illuminate how many of the eight reasons for value co-destruction are relevant to the public sector. Furthermore, the study is also based on interviews with providers, taking these providers’ viewpoints on value co-destruction into account.

Although value co-destruction might happen for the same reasons in both private and public organizations, these organizations operate under different conditions. Accordingly, this paper aims to investigate the reasons for value co-destruction in the public sector and to study the process from the perspective of the users experiencing the service. By tracking numerous incidents, as reported by users, the paper intends to provide a generic picture of the causes of value co-destruction in public services. We describe the details of our empirical study next.

Method

The paper is based on a qualitative study focusing on case management at two Swedish government agencies, the Social Insurance Agency and the Tax Agency. In particular, we studied the part of the case management process involving direct interaction between the two agencies and their users. In this dyadic interaction, other actors, e.g. the Swedish Public Employment Service, medical services, and employers, were also frequently referred to, entailing that the collected data enabled us to understand resource integration and value co-destruction in a limited service ecosystem context.

As with all agencies in Sweden’s public administration system, the Social Insurance Agency and the Tax Agency operate with great autonomy. Indeed, the direct involvement of politicians in the day-to-day running of government agencies is prohibited by law in Sweden. The agencies under study, in line with Sweden’s public sector in general, face the challenge of combining good customer service with high efficiency (Molander, Nilsson, and Schick 2002). To manage this paradox between good service and efficiency, the agencies under study have switched to providing many of their services on-line and/or by phone; there is little service provision face-to-face. Furthermore, both PSOs are characterized by bureaucratic management and administrative regimes. While these characteristics may limit our findings, they also provide opportunities for collecting data by enquiring into the resource integration process involving employees, users, and other implicated actors, as described below.

Data collection

The critical incident technique (CIT), which has been extensively used in service research (Gremler 2004), inspired the data collection used in the present study. A critical incident is understood to be an incident either contributing to or detracting from the general aim of a certain activity in a significant way (Bitner, Booms, and Tetreault 1990, 73). The CIT is commonly used in such a way that the respondent, who may be either an employee or a customer, describes the incident and its circumstances in his/her own words.

Informed by the CIT, we collected data via interviews with users who called the two PSOs included in the study. We focused on users reporting experiences of value diminishment, i.e. negative critical incidents, during the resource integration with these agencies. This design avoids the recall problem associated with the CIT approach due to respondents not being asked to report incidents happening a long time ago (Gremler 2004; Michel 2001).

The data was collected by a number of experienced caseworkers. These were given two days’ training in how to conduct interviews with users. The officers worked in pairs. They sat next to each other, each with a computer, listening to users calling in via separate headsets. Officer 1 answered the call and responded to the user’s questions according to normal procedures, while officer 2 co-listened and made notes. After the user’s issue had been addressed, they were asked to answer more questions. If s/he agreed, then officer 2 did a structured interview while officer 1 co-listened and took notes. The interview format included seven questions: (1) What attempts to access services via the PSO had the user previously made? (2) What happened during these previous attempts? (3) How did the user experience these contacts? (4) What information was obtained via the previous contacts? (5) Why was it inevitable to make new contact? (6) What information was missing? and, finally, (7) Will it be necessary to contact any other actors? While none of these questions addressed critical incidents explicitly, questions two and three were intended to capture the critical incidents experienced by the interviewees. This turned out to be the case as the interviewees frequently shared their experiences of critical incidents when answering these questions.

The officers conducting the interviews made notes regarding each interview in a web form consisting of two open text fields. In the first field, they described the critical incident reported by the user. In the second field, they noted what the client had recalled from the previous steps of the service delivery process. During the study, the researchers reviewed the collected data through the web form and were in contact with the officers doing the data collecting to make sure that it was being done properly. The data collection procedure was tested during a pilot study, leading to some minor changes in the design.

In total, 996 interviews were conducted over a period of two months. Of these, 51 were not included in the study due to being poorly described, entailing that 945 interviews were used. Of these 945 interviews, 741 were conducted with users of the Insurance Agency’s services; the remaining 204 interviews were conducted with users of the services provided by the Tax Agency. The incidents that were reported varied from user to user, allowing us to analyse the data in order to find the causes of value co-destruction and which actors – agencies, users, or third parties – are responsible for it.

Data analysis

To analyse the data, we used the coding process suggested by Gioia, Corley, and Hamilton (2013), which is based on grounded theory (Strauss and Corbin 1990). In accordance with this coding process, data was organized into first-order concepts, second-order themes, and dimensions. During analysis of the first order concepts, the researcher stayed close to the informants’ terms. In this study, first-order concepts represent descriptions of the critical incidents the informants had reported, i.e. the main reason why the agency was contacted in the first place. The second-order themes are the result of finding similarities between the first-order codes and have been generated by the researcher. The final step of the coding process of Gioia, Corley, and Hamilton (2013) is to develop overarching aggregate dimensions, which in this case represent dimensions of value co-destruction.

One important aspect of understanding value co-destruction is finding out what the true factors are which, in practice, lead to the diminishment of value for one or more of the actors in the service ecosystem. This may be difficult because the cause may not be found anywhere near the actor experiencing a diminishment of his/her value; hence a new (recovery) process needs to be initiated, i.e. there is a need for further interaction. The triggering action may often simply be the last link in a long causal chain.

The data structure illustrated in Figure 1 presents the results of the coding process; the first-order concepts, the second-order themes and the aggregate dimensions representing the causes of co-destruction, independently of which party caused it. The number of mistakes is also recorded, and listed as part of the second-order themes in Figure 1. The percentages show the cumulative distribution, across both agencies, of the apparent causes of value co-destruction. The reason for the numbers adding up to more than 100% is that several critical incidents could be assigned to more than one second-order theme.

Figure 1. Framework for coding

Data analysis of the causes of value co-destruction commenced with five researchers independently coding a selection of 50 interviews. These researchers then met in order to compare and discuss their coding. The individual coding was largely consistent between the five researchers as regards whether or not a particular incident was to be classified as a case of co-destruction. The final coding suggested that the proportion of co-destruction reported was 70 percent in the Social Insurance Agency data (of 741 interviews) and 41 percent in the Tax Agency data (of 204 interviews). However, the dimensions and the first-order concepts and second-order themes differed somewhat from researcher to researcher. Following discussions, a preliminary coding framework was developed which was further refined by the two researchers coding the remaining interviews (see Figure 1). The result of the coding did not show any significant differences between the two agencies. Despite the fact that interpretational saturation had been achieved early on, entailing that the coding of additional interviews did not have any effect on the data structure, all the interviews were coded.

The trustworthiness of the coding was secured through member checks (Lincoln and Guba 1985). The coding was presented to managers and staff at both agencies, which basically confirmed the codes arguing that they mapped their practice and were easy to comprehend. The results of the coding are presented in the first Findings section, illuminating the causes of value co-destruction.

Following the same procedure as regards coding the causes of value co-destruction, the material was also coded as regards which actor that was responsible for it – the agency, the user, or a third party. However, value co-destruction is not necessarily determined by one particular incident caused by one single actor but by multiple interacting actors. Hence, defining who is responsible for the misuse of resources can be a complex matter. For many of the incidents identified in our data, it was obvious that one single party had misused his/her own or others’ resources, causing the diminishment of value. However, other incidents were harder to trace back in order to find out which actor had been responsible. When more than one party was believed to be responsible for causing co-destruction, both parties were marked.

Findings

This section is divided into two parts. Based on the aim of the paper, the first part provides an overview of the causes of value co-destruction identified, while the second part focuses on which actors are responsible for causing value co-destruction.

Causes of value co-destruction

As shown in Figure 1, four types of causes of value co-destruction were identified in the present study: (1) A lack of transparency, (2) mistakes, (3) a lack of bureaucratic competence, and (4) an inability to serve. These dimensions represent a decrease in the wellbeing of one or more of the involved actors due to various actors’ misuse of resources.

Lack of transparency

The findings indicate that value diminishment is largely caused by a lack of transparency as regards any aspect of service provision. The dimension comprises two different aspects of transparency; first, a lack of information about a specific case and, second, a lack of information about the service ecosystem. Both of these trigger a high number of the reported incidents, 44% and 26% respectively (see Figure 1). Users make contact and seek information, for instance, concerning whether their documents have arrived or why there has been no response to their case. The response requested from an agency may involve substantive decisions or information about when a payment can be expected, which are issues of great importance to clients. Consequently, when users ‘lose sight’ of their case, due to the handling of the case, they then require more information. The lack of a transparent process is not just linked to a user’s specific case but also to the system that the user becomes part of through his/her case. Information is sometimes unclear as regards which actors are involved. One example from the collected data concerns a lack of clarity on the roles of employers, doctors and citizens with respect to decisions about sick leave. If a party is unsure about the system’s design or function, then that party will search for more information, which again may delay the handling of the case and diminish value.

Mistakes

Mistakes by PSOs, users, or third parties constitute the second type of cause of value co-destruction. Even simple operational errors, made either by persons or machines, cause diminishment of the value experienced by the user. The data material includes examples of machines making scanning errors, administrators sending out incorrect forms, and users submitting incorrect information in various ways. As seen in Figure 1, mistakes are a common cause of value diminishment, with 34% of the recorded incidents being classified as minor errors. These small and apparently quite trivial errors may have a major impact on the user’s value experience. For instance, a form sent to the wrong address can prolong the waiting time for a user hoping to receive sick-leave payments, entailing that this person may run out of financial assets. The findings also show how actors sometimes lose a document, or some other information, which is needed to pursue the case. For example, the data shows that users lose payment slips.

Lack of bureaucratic competence

The third cause of value co-destruction, a lack of bureaucratic competence, entails poor knowledge of the rules and regulations governing a case. But it also includes a more specific lack of competencies, e.g. an inability to fill in forms or not knowing how to contact a doctor. A lack of bureaucratic competence impedes the interaction between the actors involved in a case and may lead to value co-destruction in two ways, as shown in Figure 1: (1) not knowing the regulatory framework and (2) being unsure of how to navigate within the service ecosystem.

The findings show that users strive to both get information and understand it as regards the regulatory framework they are facing when interacting with agencies. For some users, this can be their first encounter with an agency and a lack of bureaucratic competence may lead to value co-destruction. The client is sometimes in need of clarification with regard to various descriptions of the rules applicable. Confusion may result from visiting a website, reading a brochure, or being informed orally by a civil servant. Information can be difficult for users to find and/or the bureaucratic language can be challenging to comprehend. As the agencies forming part of the study largely offer services online, the poor design of digitalized platforms, i.e. making them difficult to navigate and distorting information provided to users, commonly causes value co-destruction.

The findings also suggest that value is diminished on the basis of the fact that actors struggle to navigate the system; in particular, they lack knowledge of which party is responsible for the handling of their cases. There are examples of how users are referred backwards and forwards between one part of the system and another. Users initially choose, in some cases, the wrong departure point when dealing with their cases; however, the findings also show that they are being wrongly directed by other actors that are there to help them, e.g. caseworkers and experts, within the system. Some actors also lack an appropriate overview of the system, not knowing the other actors’ roles and functions. These types of ‘system-bumps’ may cause time lags in the interaction process between PSOs and users, leading to delays in payments or the receiving of other legitimate benefits.

Inability to serve

The final cause of value co-destruction is PSOs’ inability to serve. Primarily, an inability to serve has to do with users being unable to get in contact with PSOs. Specifically, this is due to caseworkers either being on leave or lengthy queues. There are also situations where technology glitches prevent users reaching PSOs and third parties with their cases. Fax machines and computers sometimes malfunction and it can sometimes be difficult to log on to an e-service. A resource does not work and a new form of interaction must be tried. In stressful situations, these types of obstacles create problems for the user. The findings also show that an inability to serve is related to changes implemented by the agencies, resulting in new applications sometimes being launched when they do not work properly. This can, for example, include a change to an e-service or a PSO website, causing problems for users which then lead to the diminishment of value.

In sum, the four causes of value co-destruction identified in the present paper represent a wide and general representation of the phenomenon. They depict value co-destruction as occurring randomly and ad-hoc, and as being the result of glitches that are a part of the service ecosystem. Mistakes and the inability to serve are commonly one-off events that represent discrete value diminishment experiences for the user. The lack of transparency and bureaucratic competence, on the other hand, represents reoccurring incidents which, for example, are built into the design of the PSOs.

Of the four causes, the inability to serve represents a dimension where the agencies bear the responsibility. The three other identified dimensions of value co-destruction, however, are caused by different parties’ intentional and/or unintentional misuse of resources. The topic we turn to next is which actors are responsible for causing value co-destruction.

Actors’ responsibility for value co-destruction

In line with previous research, the present study suggests that value co-destruction not only takes place between PSOs and users, but between multiple types of interacting actors (Plé and Chumpitaz Cáceres 2010; Prior and Marcos-Cuevas 2016). The great majority of the incidents causing value co-destruction could be traced back to one single actor. Among these incidents, the findings show that approximately half of them were caused by the agencies, that about one third were caused by the clients, and that the rest originate from a third party. However, backward tracing was less certain in other cases, indicating that more than one actor was responsible for causing the diminishment of value. These incidents were categorized as misuse by multiple actors. Consequently, this paper suggests that incidents causing value co-destruction can be divided into two main categories; i.e. the misuse of resources by a single actor or the misuse of resources by several actors. In Table 1, the two categories are illustrated using empirical examples, as reported by the caseworkers conducting the interviews with the users.

Table 1. Empirical examples of actors’ misuse of resources

Misuse of resources by single actors			Misuse of resources by multiple actors
Misuse by agency	Misuse by users	Misuse by third party	Misuse by user and agency	Misuse by user and third party	Misuse by agency and third party
A. The client believes that he/she has received too little compensation. By checking, we can see that the compensation for one period is missing. We thus make a call to the handling agent’s office for rapid processing.	C. The client’s new-born daughter was registered with an incorrect first name. When filling in the application, the client had underlined the second name in the sequence (a middle name, a common way in Sweden of marking who the person is to be known as) and thus thought that it would be recorded as a first name. We urge her to submit a new application with the names in the correct order.	E. The client tells us that her husband has been on sick leave since September and that, even now in January, he has still not received a payment. The doctor had filled out an inadequate medical certificate and, according to the wife, the husband had phoned the doctor about five times regarding this. Each time, the doctor claims it has now been done.	G. The client says that we gave her the wrong street number in the population database. We find her notification and can see that she filled in ‘1C’ but the machine read this as ‘10’.	I. The client calls because of a letter he has received from us in which we ask if he was still ill after a specific date. He has not. During the call, the officer realizes that the client applied using the wrong date, which seems to be because his employer had reported him sick from the wrong date. The employer has to contact us to establish the first day of sick leave.	K. When the client sought medical treatment for his son, it was discovered that her son was not in the register. She says that the son used to be in the register. But, since we could see that the son is in the population database, the problem must have something to do with healthcare data transmission from the population register. The client is thus referred back to the health service.
B. The client has received a letter from us to the effect that she, as stated by her employer, is now working 32 hours a week and has failed to tell us (the agency) about it. But when checked, the client had stated this in her previous application and had also been notified of a new qualifying income of 32 hours so she didn’t understand why the letter had been sent.	D. The client opposed the late fee he received regarding his employer’s declaration. He was deregistered as an employer in September and thus thought that he didn’t have to submit an employer declaration for that month.	F. The client wants to know the progress of his sick leave case and asks what is still missing in order for him to get money. We can see that his medical certificate has been received but that his employer has not yet sent in a sickness report.	H. The client calls about applying for child care using self-service and has problems filling in the application. The manager logs in as the client and helps him fill it in.	J. Social services have urged the client to notify us that his mother is ill. The client says his mother has never worked in Sweden and we tell him that, if so, she won’t be able to receive any sick pay.	L. The client has had a tax debt. This debt has now been depreciated by the Enforcement Agency. The client is calling the Tax Agency because the debt is still visible in the system. He tried to change this during an earlier call but then the Tax Agency told him to contact the Enforcement Agency, which in turn told him to contact the Tax Agency.

Table 1 provides several empirical illustrations of ‘Misuse of resources by single actors’. For example, the empirical illustrations show that the agency made mistakes with payments (example A), that an employer (third party) forgot to send in the required sickness report (example F), and that a user wrote the wrong first name for his/her new-born daughter (example C). These are all examples of incidents (previously identified as mistakes), that are triggered by a single actor’s misuse of resources, causing value diminishment.

Table 1 also empirically demonstrates ‘Misuse of resources by multiple actors’ involving the agency and the user, the user and the third party, and the agencies and the third party. Table 1 also shows that the misuse of resources by multiple actors has different reasons. First, there are incidents, like example G, which are caused by technical muddles by machines and humans (i.e. problems scanning the user’s handwriting). Second, some incidents (see, for example, H) originate from human-induced confusion, e.g. the agency poorly communicating its message and/or the recipient poorly interpreting a message, leading to misunderstandings that cause value co-destruction. Finally, the findings show that value is co-destroyed by several actors due to confusion regarding what type of roles and responsibilities the different actors have in the service ecosystem. Examples I and J illustrate that value is diminished by actors’ lack of system knowledge – one of the aspects of ‘lack of transparency’. The incidents show that actors lack an understanding of which action in the service ecosystem that caused the problem. When neither the user nor the third party has any knowledge of the different PSOs’ areas of responsibility, users may be misdirected or misinformed and then interact with an inappropriate agency, causing the misuse of resources. The crux here is who should be responsible for knowing; the user or the third party? Similar confusion regarding different actors’ roles and responsibilities in the service ecosystem also characterizes the interactions between the agencies and the third parties. Examples K and L in Table 1 both show how misunderstandings by the interacting PSOs influence the users, causing value co-destruction.

Another aspect of actors’ roles in value co-destruction is whether they intentionally or unintentionally misuse their own or other actors’ resources (Plé and Chumpitaz Cáceres 2010). The findings of the present study suggest that the bulk of co-destruction incidents result from the unintentional misuse of resources. Users, agencies and third parties all make mistakes in which there are no obvious ‘winners’. There are some actions undertaken by users that can be argued to be unnecessary behaviour, and which can be seen as intentional misuse of PSO’s resources. The incidents caused by a lack of transparency, can be said to include such behaviour. For example, some incidents in this category involve users contacting a PSO to check whether or not the agency has received submitted documents, or to check that their case is being processed. These enquiries are typically made before the response time has expired, thus sooner than the user can expect an answer. However, the users’ explain these requests with the need to gain both information and reassurance concerning their cases due to uncertainty about what is going on and suspecting that errors might have occurred. Thus, the counter argument against interpreting these user-induced incidents as the intentional misuse of resources could be that the actions of users are essential to them, and thus not intended to affect another actor’s well-being negatively.

Discussions and contributions

The notion of value co-destruction is not new in the service literature, but has been little discussed and studied in a public sector context. Therefore, the aim of this paper is twofold; i.e. to understand what the causes of value co-destruction in public services are and to understand who causes it. The paper responds to calls for research on value co-destruction in the public sector (e.g. Järvi, Kähkönen, and Torvinen 2018; Osborne, Radnor, and Strokosch 2016), contributing in a number of ways.

The paper makes an initial contribution by identifying four general causes of value co-destruction in the public sector: i.e. a lack of transparency, mistakes, a lack of bureaucratic competence, and an inability to serve. Specifically, the paper argues that these causes of value co-destruction stem from the accidental misuse of resources (e.g. Plé and Chumpitaz Cáceres 2010) during resource integration processes conducted by one or more of the actors involved in them. Two of the four identified causes of value co-destruction, i.e. mistakes and an inability to serve, are also mentioned by Järvi, Kähkönen, and Torvinen (2018), which is one of the few previous studies of value co-destruction in the public sector. Both of these causes are general, entailing that they apply to all types of actors. The other two dimensions, a lack of transparency, which accounts for the bulk of co-destruction, according to the findings, and a lack of bureaucratic competence, have not been identified in prior studies. A likely reason for the differing results is that Järvi, Kähkönen, and Torvinen (2018) studied value co-destruction from the organization’s perspective, while the present study also included the users and third parties. In addition, Järvi, Kähkönen, and Torvinen (2018) also included both public and private organizations. A lack of transparency and a lack of bureaucratic competence may be causes that are more salient in the public context.

According to the PSL framework (see, for example, Alford 2014, 2016; Osborne 2010, 2018; Radnor et al. 2014; Osborne, Radnor, and Strokosch 2016), value is co-created in interactions between the PSO and the user. In line with some service research (Echeverri and Skålén 2011; Plé and Chumpitaz Cáceres 2010), this paper contributes by showing that value may also be co-destroyed in direct interactions. Hence, the paper extends the PSL perspective by offering a more complete picture of how users enact public services, pointing to problems in the resource integration process. Previous research suggests that resource integration that leads to positive outcomes takes place in three spheres; i.e. the provider sphere, where value is facilitated by the provider of the service, the joint sphere, where value is co-created in direct interaction with other actors, and in the user sphere, where the user is an independent value creator and where value is created through the use and integration of both one’s own and others’ (e.g. the provider’s) resources (Grönroos 2011; Grönroos and Voima 2013). We argue that the spheres can also be applied when analysing disharmonious resource integration processes (Echeverri and Skålén 2011).

Value realization and fulfilment are dependent on actors being prepared for, and able to effectively make use of, the value co-creation opportunity (Grönroos and Voima 2013; Vafeas, Hughes, and Hilton 2016). However, when users do not possess the necessary resources, and/or when PSOs fail to make service offerings that their users can make sense of and employ to create value, value will be, as shown in our study, co-destructed. The agencies featured in the present study rely heavily on online services, and are largely dependent on the user’s ability to make sense of the agencies service offerings in the user’s sphere. When they fail to do so, by misjudging for instance, users’ bureaucratic competence and/or neglecting to acknowledge their need to have transparency in their cases, value will be co-destructed. The findings point to the importance of how value is always uniquely determined by the beneficiary (Vargo and Lusch 2016); in this case, users of PSOs’ value propositions, and how failing to acknowledge the users’ resources, and their use of these, may turn intended harmonious processes into disharmonious ones.

The three spheres suggested by Grönroos (2011, 2019) focus only on the user and the provider. However, as recently argued by PSL scholar Petrescu (2019), and as suggested by our study, value co-creation and co-destruction are embedded in service ecosystems where multiple actors engage in resource integration processes (Edvardsson, Tronvoll, and Gruber 2011). Addressing the part of the research question focusing on who causes value co-destruction, the findings suggest that, while the PSOs cause the bulk of co-destruction, the users and third parties also instigate it. Incidents causing value co-destruction are also likely to occur in interactions between multiple actors and to be caused by several parties. Our findings thus suggest that value co-destruction and value co-creation are multi-actor phenomena that need to be studied in service ecosystems. Our findings further substatiate the conceptual study of Plé and Chumpitaz Cáceres (2010) using empirical evidence suggesting that value co-destruction is an effect of disharmonious processes between different service ecosystem actors. The service ecosystem has become a key concept in understanding multi-actor resource integration in service research, but it has only recently been rigorously introduced into the PSL by Petrescu (2019). Our paper, together with the work of Eriksson et al. (2019), is the first to show the usefulness of the concept to PSL through empirical research.

Figure 2 serves to illustrate how both harmonious and disharmonious processes are based on resource integration between actors, set in a broader service ecosystem. Figure 2 also incorporates the link between the co-creation and the co-destruction of value. Value co-destruction is not commonly a permanent state, neither in terms of an absolute loss of value (Vafeas, Hughes, and Hilton 2016) nor in terms of time (e.g. Xu, Tronvoll, and Edvardsson 2014). In the public context, the recovery of value diminishment is a particular imperative. Compared to customers of private organizations, users of PSOs rarely have any alternative opportunities; for instance, there is no ‘escape’ from the Tax Agency. Despite offering a service ecosystem view of value co-destruction, we argue, in line with extensive public management research (Alford 2016; Osborne 2010, 2018; Osborne, Radnor, and Strokosch 2016; Sørensen and Torfing 2011), that PSOs have formal authority and are thus the actors most clearly responsible for ensuring and developing strategies for reducing the causes of value co-destruction and for turning value loss into value gain. The PSO represents the value facilitator, providing value propositions, with the users, jointly with other actors and resources, co-creating value (Eriksson et al. 2019; Petrescu 2019; Skålén et al. 2018). PSOs are also formally accountable when it comes to addressing causes of value diminishment. Hence, it is important that the value proposition of the PSOs provides clients and other parties with the preconditions to facilitate the co-creation rather than the co-destruction of value.

Figure 2. Value co-creation and value co-destruction in service ecosystems

Limitations and future research

The present paper suffers from limitations, which call for further research. Firstly, it draws on a study of value co-destruction at two similar Swedish PSOs. Although the identified causes of value co-destruction might apply in other contexts, there is still a need for more studies covering the diversity of public service contexts. Such future studies need to quantitatively investigate whether or not the findings pertaining to the causes of value co-destruction, and which actors are responsible for it, outlined here, hold across contexts. In addition, future studies also need to collect richer qualitative data than the type of data reported on here, which was collected by public officials, in order to gain a more in-depth understanding of value co-destruction in a public context. Furthermore, the paper also focuses on a limited service ecosystem, namely PSOs, the users of services, and third parties; i.e. actors directly involved in value co-creation. In order to come to terms with value co-creation and value co-destruction within the PSL framework, a bigger proportion of the key actors comprising the public sector needs to be included. Such studies may contribute to our understanding of the public sector from a service ecosystem perspective and may also further contribute towards imbuing the PSL with service ecosystems thinking. Finally, since the paper draws on users’ own experiences of value diminishment, its main contribution is linked to the co-destruction of individual value – not the diminishment of group value or public value (e.g. Alford 2016; Moore 1995; Williams, Kang, and Johnson 2016). Following Alford (2016), there is an interdependence between private value and public value. Further research would likely contribute by studying value as individual and collective phenomena.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Marit Engen

Marit Engen is a Post.doc at the Service Research Center, Karlstad University, Sweden. Marit is working within the domains of service research, service innovation and employee-driven innovation. She is currently conducting empirical research on the opportunities and constraints of using service design in the development of services aimed at vulnerable groups. Her work has appeared in journals including: The Service Industries Journal, Creativity and Innovation Management, Australian Journal of Public Administration and Journal of Quality Assurance in Hospitality & Tourism.

Martin Fransson

Martin Fransson is an Associate Professor at Service Research Center, Karlstad University. Martin has a PhD in Business Administration and the focus of his research is within public service logic. It involves management and control as well as service design, organizing and various topics within public administration. At the moment Martin is hired as a Research Manager by the Swedish government.

Johan Quist

Johan Quist is an Associate Professor at Service Research Center, Karlstad University. Johan’s primary research interest is the encounter between citizens and public sector organizations. He is also, together with Martin Fransson, Head of research for a Committee working for the Swedish Government and with a task to improve public governance and management. The Committee is called Delegation for Trust-based Public Management, reporting to the Swedish Minister of Public Administration.

Per Skålén

Per Skålén is a Professor of Business Administration based at the Service Research Center, Karlstad University, Sweden. Per is currently working within the domains of service research, public management research, critical marketing, and with applying practice theory, strategic action field theory and other sociological theories to marketing and consumer research. He is currently conducting empirical research about the transformation of the Swedish music industry and about marketing work. His work has appeared in several journals including: Journal of the Academy of Marketing Science, Journal of Service Research, Marketing Theory, Journal of Marketing Management, Journal of Public Policy & Marketing, Organization, and Tourism Management.