If digitalization is the answer – what is the problem? An analysis of policy documents related to the digitalization of Norwegian child welfare services

ABSTRACT

In this article, we have analysed policy documents related to the digitalization of the Norwegian child welfare services. The project ‘DigiBarnevern’ is a national initiative intended to develop and improve the digital tools in child welfare services. In our article, we explore which arguments for digitalizing child welfare are used in Norwegian policy documents. The analysis is inspired by Carol Bacchi’s analytical approach of ‘What’s the problem represented to be?’ This approach starts with the premise that policy documents are produced and implemented as answers to specific problems. Our analysis shows how the problems represented are inefficient services, a lack of systematization, arbitrary practices, and the lack of a quality system. Thus, digitalization seems to be the answer to comprehensive problems in child welfare services. By drawing on the literature of street-level bureaucracy and professional discretion, we discuss features of digital systems such as their potential to measure quality and how they can serve as tools to meet the aim of regulating frontline workers` room for discretion. Policy documents are not neutral tools. Instead, they are instruments and carriers of values that affect service provision and ultimately the clients.

SAMMENDRAG

I denne artikkelen har vi analysert politiske styringsdokumenter knyttet til digitaliseringen av det norske barnevernet. Digitaliseringsprosjektet «DigiBarnevern» er et nasjonalt initiativ som har som formål å utvikle og forbedre de digitale løsningene i barnevernet. I vår artikkel utforsker vi hvilke argumenter for digitalisering av barnevernet som brukes i norske politiske styringsdokumenter. Analysen er inspirert av Carol Bacchis analytiske tilnærming til ‘What’s the problem represented to be?’ Denne tilnærmingen starter med premisset om at politiske styringsdokumenter produseres og implementeres som svar på spesifikke problemer. Vår analyse viser hvordan problemene representert er ineffektive tjenester, mangel på systematisering, vilkårlig praksis og mangel på et kvalitetssystem. Digitalisering ser altså ut til å være svaret på omfattende utfordringer i barnevernet. Ved å trekke på litteraturen om Michael Lipskys (2010) bakkebyråkrati og profesjonelt skjønn, diskuterer vi trekk ved digitale systemer og deres potensiale til å måle kvalitet, samt hvordan de kan tjene som verktøy for å møte målet om å redusere ansattes handlingsrom. Politiske dokumenter er ikke nøytrale verktøy. I stedet er de instrumenter og bærere av verdier som påvirker tjenestetilbudet, og ikke minst, tjenestemottakerne.

KEYWORDS:

• Child welfare
• digitalization
• standard
• street-level bureaucracy
• discretion

NØKKELORD:

• Barnevern
• digitalisering
• standardisering
• bakkebyråkrati
• skjønnsutøvelse

Introduction

In this article, we explore which problems digitalization is meant to solve within Norwegian child welfare services. Digitalization of the public sector in Norway has been on the agenda for many years, but few state-initiated solutions have been introduced into child welfare services, compared to other parts of the public sector (Hansen, Lundberg, and Syltevik 2018). The expectations of both service users and frontline professionals are said to rise in line with the increased use of digital and social media in society (López Peláez et al., 2018). These demands have led to greater use of digital and social media in child welfare practices to meet clients’ needs for efficient and accessible forms of communication (Kvakic, Fineide, and Hansen 2021). In recent years, the initiation of the digitalization project ‘DigiBarnevern’ (Digi Child Welfare) has been one attempt to meet some of the new demands with better and more accessible digital services. The project aims to develop and implement several new digital solutions, including a standardized case assessment tool to ensure certain documentation standards, more equal service provision and increased user participation. Not all the DigiBarnevern solutions have been implemented to date (2023), but they are described in various policy documents.

Recent studies show how professionals` digital and social media use challenges them in new ways. Examples are professionals having different approaches and coping strategies for their feeling of constantly being available for clients, increased documentation, online client surveillance and even a potential loss of trust in professionals when they search for client information on social media platforms (Kvakic and Wærdahl 2022; Kvakic, Fineide, and Hansen 2021). Since the Norwegian State is aiming to digitalize public services, including child welfare, we wish to find out which arguments for digitalizing child welfare are used in Norwegian policy documents.

Digitalization can be defined in various ways. It can, for example, be referred to as processes of automation and shifting from manual to digital processes (Ask and Søraa, 2021). One example is shifting from paper-based mail and face-to-face meetings to digital mail and digital meetings. As a government initiative, digitalizing public services is often said to lead to cost savings, increased efficiency and quality improvement. Ask and Søraa (2021) refer to this rhetoric as positioning digitalization as a ‘technological miracle cure’ with the aim of solving a wide variety of societal problems (Ask and Søraa 2021, p. 29). In this paper, we refer to digitalization as the overall shift towards increased implementation and use of various digital systems and tools, such as ICT (information and communication technology) systems for communication and casework in child welfare services.

We acknowledge that policy documents are among the most influential tools of power (Lundgren et al. 2012). According to Lascoumes and Le Gales (2007), public policy instrumentation implies a set of problems posed by the choice and use of instruments that allow government policy to be made material and operational. Policy instruments can be seen as ‘bearers of values, fuelled by one interpretation of the social and by precise notions of the mode of the regulation envisaged’ (Lascoumes and Le Gales 2007, 4). Thus, the question of which interests are involved in the choice of instruments is both important and necessary (Lascoumes and Le Gales 2007). The rationale for using policy documents as an empirical window to study digitalization in child welfare services is to provide insight into the relationship between those governing and those being governed. By studying policy documents, this article explores the arguments used in policy documents that present digitalization as the solution to complex challenges in child welfare.

The organization and role of the Norwegian child welfare services

The Norwegian child welfare services are organized at municipal and State levels. Frontline workers are in direct contact with clients on the municipal level. Here, children’s care situations are investigated and appropriate interventions are implemented and evaluated by caseworkers. The child welfare services’ mandate is to safeguard children’s right to a secure childhood environment and to provide the necessary help and care to those children who live under conditions that may harm their health and/or development (Child Welfare Act 2021). The role of child welfare is to aid families and children when there is concern about a child’s care situation. Professionals assess the care situation of children and youth, implement and evaluate interventions, and exercise coercion when considered necessary. The exercise of coercion is part of the child welfare’s mandate, and the Child Welfare Act gives municipal services the possibility to move children out of their home, based on a court ruling. This places great demands on child welfare services` work, including to ensure that the legal rights of children and parents are being adequately protected (Ministry of Children and Families 2021). In general, municipalities are at great liberty to decide on their preferred form of organization of services. Their choices are therefore based on local circumstances and interpretations of managerial ambitions (Rønhovde 2016, 80).

The main tasks of the state part of child welfare, the Office for Children, Youth and Family Affairs (Bufetat), are to run state care facilities and to help municipal child welfare to find appropriate out of home placements for children and youth in need of such services (Directorate for Children, Youth and Family Affairs 2021).

Child welfare services’ various challenges and attempts in solving them

The Norwegian State and child welfare seems to be facing a number of challenges: convictions for violating human rights in several child welfare cases (Norwegian National Human Rights Institution 2021), lack of evaluation and documentation of care interventions and decisions on care orders (Ministry of Children and Families 2021; Norwegian Board of Health Supervision, 2019), and inadequacies in conducting and assessing cases and ensuring children’s needs for physical, emotional and legal protection (Official Norwegian Report, 2017). One concern raised is that investigations conducted in child welfare are largely based on discretion, rather than structural procedures (Samsonsen 2016).

Some of the described challenges are intended to be met through the digitalization project DigiBarnevern (Norwegian Association of Local and Regional Authorities 2018). The project is divided into two parts: one for municipal services and one for state services. For the municipal part, the Norwegian Association of Local and Regional Authorities (KS) and eight of the largest municipalities in Norway are to develop and implement Citizen Services, a digital portal where citizens and child welfare professionals can communicate through a digital channel that safeguards privacy and confidentiality. The portal will also give child welfare clients access to their documents and the assessment process of their case. The municipalities were also given the responsibility for reviewing the need for a new ICT system and inviting tenders to develop ICT solutions that could provide professionals with better support than the existing ICT systems on the market (Norwegian Association of Local and Regional Authorities 2018). ICT in this context means digital systems for communicating with clients, documentation and digital case assessment tools. These systems seem to be introduced as top-down initiatives from the authorities and/or managers, as opposed to SMS, email and social media platforms, which are often used by the caseworkers themselves (Kvakic, Fineide, and Hansen 2021).

The Directorate for Children, Youth and Family Affairs (Bufdir) is responsible for the state part of the project. The Directorate has developed a new, common case assessment model called the Child Welfare Quality System and a digital portal for sending and receiving notifications of concern about children. Additionally, the Directorate is developing standards and models for reporting and registering child welfare data with the help of the new ICT systems, providing authorities with easier and better access to information. This information is stated to be used by the state authorities for statistics and analysis for knowledge-building purposes (Directorate for Children, Youth and Family Affairs, 2020).

Before the DigiBarnevern initiative, there was no national, standard model for child welfare investigations. With the DigiBarnevern initiative, however, the State seeks to standardize investigations with a new common assessment and investigation model (Ministry of Children and Families 2021).

On top of the issues mentioned above, high work pressure, psychologically demanding cases and low salaries are contributing factors to high turnover. A recent study revealed that four out of ten caseworkers are actively searching for another job and planning to leave their current position in the child welfare within the next year (Directorate for Children, Youth and Family Affairs 2022).

Another attempt at meeting some of the challenges in child welfare is through the revision of the Child Welfare Act, which includes an increase in the documentation of and justification for the reasoning behind various decisions (Ministry of Children and Families 2021).

While digital services and ICT tools can be an advantage for clients possessing the necessary digital skills and experience, they can be a disadvantage and even exclude citizens with lower socio-economic status, lower education and poor language skills (Breit and Salomon 2015; Buchert et al. 2022; Hansen, Lundberg, and Syltevik 2018; Pollitt 2010). Some critical voices point to ICT systems as reductionist, being unable to capture the ‘whole picture’ of a client’s situation (Brodkin, 2013; Devlieghere, 2019; Gillingham 2019). Thus, the ICT systems` potential to serve as accountability tools and quality measurers become limited to the categories and features predefined by their developers. Studies show how professionals find ways to circumvent ICT systems they find restricting (Devlieghere and Roose 2018; Røhnebæk 2016). These studies demonstrate ICT tools` limitations in steering and controlling professionals` discretionary space and service provision.

Street-level bureaucrats and their exercise of discretion

Many of the challenges the child welfare is facing can be identified in the street-level bureaucracy literature and Michael Lipsky’s (2010) work. According to Lipsky (2010), child welfare professionals, like other frontline workers serving the State, such as social workers, teachers and police officers, are considered street-level bureaucrats. What characterizes street-level bureaucrats is meeting citizens with complex needs, working under demanding conditions with limited resources and having considerable room for discretion (Lipsky 2010).

Discretion, the cornerstone of professional work, has two dimensions: a structural dimension of ‘discretionary space’ and an epistemic dimension of ‘discretionary reasoning’. The discretionary space is like the ‘hole in a doughnut’ where the circle of the doughnut comprises the ‘belt of restriction’ and where the hole in the middle may be large or small (Dworkin 1978, 31). Discretionary reasoning is defined as the cognitive activity that may take place within the discretionary space of professional judgement (Wallander and Molander 2014).

By applying our theoretical framework of street-level bureaucracy and professional discretion, we discuss what is left unproblematized and which effects are produced by the problem representation in the policy documents.

Methods

Design

The aim of our study is to critically examine and determine which problems digitalization is meant to solve. We raise the question: What arguments are used in policy documents that present digitalization as the solution to complex challenges in child welfare? The design of this study is inspired and guided by Carol Bacchi’s analytical framework ‘What’s the problem represented to be?’ (WPR) (Bacchi 2012). Bacchi’s framework is based on the premise that what one seeks to do reveals what one thinks is problematic or needs to change. Bacchi states: ‘(…) the “WPR” approach serves as a much-needed interruption to the presumption that “problems” are fixed and uncontroversial starting points for policy development’ (Bacchi 2012, 23). Policy documents always contain a specific way of continuing the problem and are created through specific representations and underlying assumptions about what the problem is represented to be, even when the problem representation is not explicitly expressed in the document.

Bacchi’s framework consists of six analytical questions. She states that it is to be conceived as an open-ended mode of critical engagement, rather than a step-by-step formula (Bacchi 2012). We chose to use four of Bacchi’s six questions, namely 1) What’s the problem represented to be in a specific policy (proposal)? 2) What presuppositions or assumptions underpin the representation of the ‘problem’? 4) What is left unproblematic in this problem representation? and 5) What effects are produced by this representation of the ‘problem’? Questions 3 and 6 ask how the representation of the ‘problem’ has come about, and how it is presented and defended. We chose to exclude questions 3 and 6 to delimit our study.

To explore our research question, we applied Bacchi’s questions 1 and 2 in the initial coding. These two questions helped to deepen our understanding of how the problems are described and what purpose they are meant to fulfil, because this can give us insight into the methods of solving the problems. Further, we were inspired by how Høyem et al. (2018) applied Bacchi’s framework in their article and document analysis regarding coordinators in Norwegian hospitals. In line with Høyem et al. (2018), we found the application of Bacchi’s question 1 and 2 in the result section, and question 4 and 5 in the discussion as a fruitful approach to structure and delimit our study.

We see policy documents as political instruments with the purpose of steering actors` thinking and acting in a way that they would otherwise not consider. This view allows us to deconstruct policies in a manner that enables us to address dimensions of public policy that would otherwise not be visible (Lascoumes and Le Gales 2007). Lascoumes and Le Gales` (2007) approach, together with Bacchi’s framework, allows us to critically interrogate public policies.

Material

We started by reading the description of requirements for DigiBarnevern, a document written by KS and the municipalities, describing the needs child welfare workers have for the new digital solution, together with a description of the State’s role in the project. The description of requirements led us to a report written by Bufdir, where an investigation into the need for a national digital system was ordered earlier that year by the Ministry of Children, Equality and Inclusion in a letter of allocation. Thus, we worked backwards from the proposed solution (digitalization) to the problem representations and descriptions (Bacchi and Goodwin 2016). Being unable to find any additional documents related to DigiBarnevern or mentioning the digitalization of child welfare, we conducted a search on the Bufdir and KS websites for any additional documents on DigiBarnevern. This led us to the DigiBarnevern final report, which described the concept phase of the DigiBarnevern project. We chose this document to find out which final digital solutions and initiatives were considered necessary by Bufdir. The final report referred to the State’s general investment in digitalizing public services, and how the aims of the DigiBarnevern project followed the relevant policy documents.

To understand the societal, political and social context of digitalization of child welfare, we performed a search on the Government’s website (regjeringen.no), where we found several documents, including two government letters and one strategy document related to digitalization of public services in general, and one proposition and two state budgets where digitalization of child welfare and/or DigiBarnevern were specifically mentioned.

Finally, we obtained 27 documents, mainly reports, prepositions, letters and documents dealing with the state budget. These were state budgets and other Government documents dealing with overlapping themes or digitalization that was not relevant for this study, such as general digitalization initiatives. A close reading of these documents enabled us to form a rough idea of the content of the text and exclude documents not relevant to our analysis. The result of this selection process was ten documents consisting of four reports, three letters, one statutory proposition and two documents dealing with the state budget, for the period 2016–2021 (see Table 1). These documents helped us to understand the background of the project, how DigiBarnevern came about, who initiated the project and why.

Table 1. Table of included documents in chronological order.

	Document title and which parts of the documents were analysed	Type and source	Topic covered
1	Investigation of a national digital system and coordinated digital management in child protection, Section 4.1.3 in letter of allocation 2016 (pp. 4–32)	Report written by the Directorate for Children, Youth and Family Affairs (Bufdir)	Digitalization of child welfare services	2016
2	Letter of allocation to Bufdir 2016 (Section 2.2, pp. 3–4, Section 4, pp. 7–8, Section 854, p. 16, Section 5.3, p. 24, Section 6.2, p. 26)	Letter of allocation, Ministry of Children, Equality and Inclusion	Digitalization of child welfare services	2016
3	Prop. 73 L (2016–2017) Changes in the Child Welfare Act (child welfare reform) (Section 1.2, p. 8, Section 6.3.2, p. 45, Section 7.2.2, pp. 48–49, Box 8.1, p. 58, Section 9.1.2, p. 65, Section 9.2.2, p. 66, Section 17.1.1, pp. 157–158)	Legislative proposition, Government	Digitalization of child welfare services	2016
4	Final report on digital citizen services and case management support in child welfare. Concept phase, version 1.00 (pp. 6–14, 24, 39–45, 66–67)	Final report on the DigiBarnevern concept phase, DigiBarnevern	DigiBarnevern	2017
5	DigiBarnevern: Description of requirements for municipal child welfare solutions, version 0.7 (pp. 4–11, 25–35, 37–53)	Description of requirements for the DigiBarnevern project, KS	DigiBarnevern	2018
6	Digitalization in the municipal sector (pp. 1–8)	Government letter	Digitalization of public services	2018
7	Starting a digital initiative for child welfare (State Budget, p. 28)	Press release (State Budget 2019)	DigiBarnevern	2018
8	One digital public sector -Digitalization strategy for the public sector 2019–2025 (Chapters 1 & 2, pp. 4–12, Chapter 3, p. 18, Chapters 4 & 5, pp. 26–37, Chapter 6, pp. 38–42, Chapter 8, pp. 46–48)	State strategy document, Ministry of Local Government and Modernization	Digitalization of public services	2019
9	Prop. 1 S (2020–2021) State Budget 2021 (pp. 17–18, 92, 95, 104, 106, 113–114, 119, 170)	State Budget	DigiBarnevern	2020
10	Digitalization in the public sector 2021 (pp. 1–6, 8–10, 12–13)	Government letter	Digitalization of public services	2021

Analysis

The first documents were read and coded by all three authors together, to ensure that we had a common understanding and application of Bacchi’s framework. The authors read and coded other texts separately before meeting and discussing findings. This meant that all documents were read by all authors, either together or separately. During the reading of the documents, each of the authors wrote memos of initial analytical thoughts, ideas and reflections. The texts were read with Bacchi’s questions 1 and 2 in mind: How are the problems described, and which arguments and presuppositions are used to underpin the problems? This resulted in condensed data, where we extracted elements from the full-text documents that were relevant for further analysis. At this point, we decided to go back for a new round of reading the extracts in the original documents. This was to ensure that we had not missed any other parts relevant for answering our research and analytical questions.

During our reading and re-reading of the documents and extracts, each of us coded the data, and we then met to discuss our coding. The discussions, coding and analytical memos led us to a number of relevant categories. After discussing the categories, we discovered different overarching themes: arbitrary practices, poor quality and inefficient services. In Table 2 is an example of the process that demonstrate how we analysed the documents, and how our codes resulted in categories and themes:

Table 2. Example of coding and categorization.

Document number	What`s the problem represented to be?	What presuppositions or assumptions underpin this representation of the ‘problem’?	Code	Category	Theme
1	Lack of central guidelines and large local variations (p. 5)	The Office of the Auditor General in Norway has argued that there is excessive room for discretion in child welfare services. (p. 5)	Too much room for discretion	Unequal treatment	Arbitrary practices

During our discussions of the themes relevant to answering our research question, we found that the problems of arbitrary practices and quality were closely linked to the need for a quality system in child welfare. We therefore decided to combine these into one theme. The problems of inefficient services were linked to a lack of systematic documentation, reporting and management. This resulted in two themes or problem descriptions: 1) inefficient services and lack of systematization, 2) arbitrary practices and lack of a quality system. These problem descriptions are closely connected and cannot be seen as independent of each other. Both are linked to descriptions of overall poor quality in child welfare and the need to strengthen the authorities` control over child welfare services. Quality deficiencies as a theme will therefore occur in both problem descriptions.

In the following, we will present our results through the two problem descriptions analysed. Questions 4 and 5 are used to explore what is left unsaid or unproblematized in the policy documents, and which effects are produced when the problems are represented. These two questions will be answered in the discussion.

Results

Problem 1: Inefficient services and lack of systematization

Quality and efficiency were closely linked in the policy documents studied. Manual processes, a lack of digital systems to speed up case processing, a lack of safe communication channels and a lack of user-friendly ICT systems were all said to lead to poor quality in service provision. There seemed to be high expectations for the new digital systems in terms of the various quality issues in child welfare that they could solve. The notion appeared to be that effectiveness and ICT systems are equivalent to good quality in service provision:

The IT solutions support modern and innovative working methods, leading to efficiency in everyday work and increased quality of services, by providing an overview and user-friendly [digital] solutions. (5, p. 41)

As we can see from this quote, the expectation seemed to be that DigiBarnevern and more efficient services would enhance the overall quality of child welfare services; quicker communication would lead to quicker service provision and support interventions that would enable children to receive the help they needed faster. The DigiBarnevern solutions were also described as freeing caseworkers from manual work processes, and providing more time for talking to children. Lack of documentation of interviews with children was one of the criticisms directed at the child welfare service, as stated in the state budget:

Quality challenges in the child welfare services are well documented: lack of systematic investigations, lack of documentation, lack of child participation, and lack of management and leadership. Case documents are sent by mail. Three out of four municipalities answer that it takes more than three weeks to get hold of all necessary documentation when a child (…) moves to a new municipality. (7)

As we can see from this extract, there are several problem descriptions. The descriptions of lack of documentation, management, systematic work and user participation are all used as arguments for why child welfare needs to be digitalized. To create systems to solve the issue of inefficient services and to accommodate interviews with children by decreasing case processing time could be one attempt to meet some of the criticism. Arguments such as ‘more time for talking to children’ appeal to caseworkers with their feelings of inadequacy because of high work pressure and lack of time. Like many other street-level bureaucracies, child welfare is subject to high turnover and burned out professionals. This issue was also said to be caused by outdated digital systems:

Complicated and user-unfriendly tools can increase the risk of sick leave and high turnover.

(5, p. 9)

The problem is here represented by outdated digital systems. The presented solution is new, user-friendly systems that can increase efficiency, prevent resources being used to train new employees to use the ICT systems, prevent delays in case processing and generally improve working conditions for child welfare professionals.

So-called ‘snail mail’ and the lack of systems for quick written communication was another problem leading to inefficient services. This applied not only to communicating with service users but also with other public services and stakeholders. The strategy document for public sector digitalization referred to an OECD report stating that even though Norway is among the most digitalized countries, there is a need to strengthen governance and coordination of digitalization efforts. The State must therefore take control when new digital solutions are to be developed in public services. The lack of coordination was not only present in the development of new digital services but also in child welfare service provision:

In particular, the project found that quality could be greatly improved in services, efficiency in case processing and interaction between child welfare, service users and other stakeholders. The project concluded that there are no ICT solutions available on the market that can cover the needs of municipal services, and that new developments [ICT tools] are necessary. (4, p. 8)

This quote demonstrates the apparent expectation that digitalization would not only improve quality and efficiency in child welfare but also generally enhance collaboration with service users and external stakeholders. The problem of poor collaboration between public services and their stakeholders has also been documented (8). The strategy document mentions a lack of incentives and cross-sectoral service provision, and how such incentives can improve cross-sectoral collaboration. This suggests that the State is attempting to meet the challenges of poor collaboration in child welfare by using DigiBarnevern to connect collaboration to digital communication and equate greater digitalization with improved collaboration.

In this way, ICT systems could be used to measure performance, such as the breadth and frequency of collaboration to improve quality and efficiency. Quality was also stated to be measurable with the help of digital systems and quality indicators (3). There was a proposal to meet the need for improved quality through digital systems and quality indicators developed and directed by the State:

Bufdir is working to develop better digital solutions for the acquisition and publication of semi-annual reports at the municipal level in order to e.g. make information more available for municipal authorities and other stakeholders. The Directorate’s [Bufdir’s] work on developing quality indicators for child welfare is also important for future reporting to make the quality of work more visible than it is today. (3, p. 45)

The problem of inefficient services lacking quality and management is here supported by the need for the State to improve organization and take control. The need for stronger central leadership is also described in the strategy document. This need is underlined with reference to a report written by the Office of the Auditor General, outlining the problem of lack of state control in the digitalization of public services. New digital systems can support the State in its aim for increased control, as demonstrated by this extract from the state budget for 2021:

(…) the state part of DigiBarnevern will prioritize the further development of a child welfare quality system and establish common solutions for reporting key information from municipal child welfare. This will replace today’s solutions for semi-annual reports from the municipalities. (9, p. 170)

The recipient of the key information reported by child welfare is not specified. In other documents, lack of management and control in municipal child welfare are specifically mentioned, such as the fact that DigiBarnevern will improve the guidelines for child welfare service managers and municipal authorities (3, 4, 7, 9). The semi-annual reports mentioned at the end of the extract are usually reported to the county governor. The county governor’s role is partly to control and inspect public services such as child welfare to ensure that they follow laws and regulations, such as the Child Welfare Act and other guidelines set by the state authorities. Thus, the digitalization of reporting and the development of digital quality indicators have the potential to give the State broader and quicker insight into efficiency (or the lack thereof) and, according to the policy documents, quality of service provision.

Problem 2: Arbitrary practices and lack of a quality system

The State’s need for control over a service dominated by many intricate challenges was also demonstrated by issues of arbitrary practices and lack of standards. As in the previous problem description, these issues were said to lead to poor quality of service provision. One problem described was the wide variation in how different managers assessed similar notifications of concern about a child. The document written by Bufdir refers to the report by the Auditor General emphasizing the need to strengthen Bufdir’s involvement in the child welfare services to ensure more equal service provision. The same report was also used to underscore how there is too much room for discretion in child welfare. The proposed legislation Prop. 73 L also indicates the problems of arbitrary practices and service provision by referring to a report by the Norwegian Board of Health Supervision. According to several documents, ICT systems will enable more equal service provision. The state budget for 2021 is one example:

DigiBarnevern will give child welfare employees professional support in conducting fair assessments and will facilitate the documentation of assessments. (…) The work on DigiBarnevern and Bufdir’s professional development work are important contributions to improve the quality of child welfare and counteract undesirable variations between child welfare services. (9, p. 92)

As this quote shows, DigiBarnevern was considered important in improving quality and ensuring equal service provision in child welfare. One of the reasons for unequal service provision was the use of different templates and tools, not only between different services but also between professionals in the same service. Because of issues with unequal and arbitrary service provision, Bufdir was tasked by the state authorities in 2017 to develop a quality system in the form of a professional framework to be used in all child welfare services in Norway. The development of the quality system is part of state responsibility in DigiBarnevern. This ensures that the State has full control in developing and maintaining the new national professional child welfare framework:

The child welfare quality system is a framework that will give municipal child welfare services systematic and knowledge-based professional support for their work processes. The professional content of the quality system will at all times reflect ‘best practice’ in child welfare and be an integral part of the digital child welfare solutions. The child welfare quality system will consist of a knowledge model, check lists, support texts, templates and more comprehensive and detailed texts. The quality system will be maintained from a central [state] point and made available for use in the municipal child welfare [digital] solutions. (5, p. 3)

DigiBarnevern and the quality system were thus presented as a solution to the problems of lack of standards, arbitrariness and excessive room for discretion. Several documents state that DigiBarnevern will support the implementation of a framework developed by the state that is expected to mitigate many of the problems caused by arbitrary practices and the lack of standards (3, 5, 7, 9).

The quality system was intended to guide caseworkers in the entire work process from receiving a notification of concern about a child until decision-making and beyond:

(…) The purpose of the quality system is to ensure good professional quality, and to provide direction for critical ‘pathway decisions’ in case processes. The system is intended to lead to more equal and quality assured child welfare practice in child welfare services. The quality system provides professional guidance for all the work processes of caseworkers in a child welfare case: notifications of concern are received, investigations initiated, and assessments and decisions are made, implemented, followed up and evaluated. (4, p. 39)

As this excerpt shows, the need of the State to control work processes and decision-making in child welfare is very clearly expressed in the policy documents. One example is how the State will develop, maintain and guide the child welfare professional framework. Another is the language used throughout the policy documents, such as how the quality system is meant to ‘guide’ and ‘support’ caseworkers in every stage of their work process. Here, the problem representation lies in lack of standards, unequal practices and arbitrariness. The presupposition is that DigiBarnevern with its digital tools, standardization and the quality system will solve many of the described problems in child welfare. According to the policy documents, the digital solutions will raise the overall quality of child welfare services, and restore the trust child welfare lacks among citizens today.

Discussion

In this article, we have analysed policy documents related to the digitalization of the Norwegian child welfare services. The primary question to be answered was what arguments are used in policy documents that present digitalization as the solution to complex challenges in child welfare. Inspired and guided by Carol Bacchi’s analytical framework, our findings indicate that quality and efficiency were closely linked in the analysed policy documents. Manual processes, lack of safe communication channels, inefficient and user-unfriendly ICT systems were all said to lead to poor quality service provision. The expectations for digital systems to solve various quality shortcomings in child welfare seemed to be high, and the notion appeared to be that efficacy and ICT systems were equivalent to high-quality service provision. Intricate challenges were demonstrated by issues such as arbitrary practice and lack of standards. As a result, increased state control seemed to be the remedy for straightening things out in child welfare services. The documents demonstrate the need for strengthening Bufdir’s involvement to ensure equal service provision to mitigate the problem of arbitrary practice and raise the overall quality of child welfare through digital means.

Our findings, where we answered how the problems are described, and which arguments and presuppositions are used to underpin the problems (questions 1 and 2 of Bacchi’s framework), represent a stepping stone for the further discussion. In the following, we will discuss what is left unsaid (question 4), and which effects are produced by this representation of the problem (question 5).

The ability of ICT to measure quality

The suggested solutions to the problem representations in the policy documents analysed were highly ambitious. Digitalization seemed to be the answer to comprehensive problems in child welfare, such as deficiencies in quality, efficiency, systematization, documentation and management, arbitrary practices, high turnover and even sickness absences. Such rhetoric accompanying the implementation of various reforms, such as the DigiBarnevern project that changes the delivery of organizational policy, is often said to make things work better by improving efficiency, coordination and accountability (Ask & Søraa, 2021; Lipsky 2010). In the policy documents, we can observe an attempt to steer services towards using digital systems to enhance performance and quality measurement. One result of this is that performance measurement ends up serving as an instrument for implementing policies (Carlstedt and Jacobsson 2017; Pollitt 2010), and selectively determining which aspects of policy matter, and which do not (Brodkin, 2013). Despite this, the authorities still aim to control what they define as quality and how welfare services should perform their practice, through e.g. supervision and various digital performance measurement tools (Engebretsen and Heggen 2012), as highlighted in our analysis.

The documents analysed do not reveal any attempts to discuss what high-quality service provision in child welfare in fact is, what it means or how, if at all, it can be measured using digital systems. When quality is measured through supervision, Engebretsen and Heggen (2012) argue that knowledge is often linked to quantity. Numbers and statistics facilitate trust, giving the reader a signal that true and valid knowledge is being conveyed. The guiding knowledge policy thus steers in a direction where exact, measurable knowledge is linked to high-quality service provision.

Brodkin (2013) stresses that when work performance is reduced to measurable categories and numbers, which is very much the case for ICT systems, some policy aims will indirectly be privileged over others. This has been highlighted as problematic by several scholars. The limited data an information system can provide makes accountability bureaucratic and limits the representation of ‘reality’ (Devlieghere, Bradt, and Roose 2019). Systems that fail to capture rationalities informing professional decision-making outside predefined categories in the ICT systems, such as local, moral, tacit and emotional factors, may at worst actually increase the risk to children’s wellbeing (Pithouse et al. 2012), because important aspects providing information about a child’s care situation may remain unnoticed (Gillingham 2019). The structural dimension of discretion, or the ‘doughnut’ that sets the boundaries for professional discretion (Dworkin 1978), may not be able to capture the ‘informal’ use of discretion, as described by Pithouse et al. (2012). As a result, the aims to restore the quality and trust in a profession by imposing standards and ICT systems on frontline professionals may, in the worst case, lead to the opposite.

Decreased scope of street-level bureaucracy

While wide room for discretion is necessary for street-level bureaucrats to meet political goals such as individually tailored services, this also implies that the state loses some of its control over professional practice (Lipsky 2010). However, when wide room for discretion leads to similar cases being assessed differently by different child welfare services, it poses challenges for the authorities. To remedy ‘undesirable variations’, reports highlight the need for increased standardization through digitalization, aiming to rectify problems of quality, inefficiency, coordination and unequal service provision (Directorate for Children, Youth and Family Affairs 2016; Norwegian Association of Local and Regional Authorities 2018). As the child welfare lacked a common framework for investigation and assessment, a quality system, as our findings suggest, could be an attempt for increased state control over professional service provision, thus aiming to restrict a profession which by nature is difficult to control (Lipsky 2010).

The implementation of digital technology by policymakers, including imposing routines on social workers, is considered an important shift, since social workers previously had considerable autonomy (Mackrill and Ebsen 2018). When a profession has legitimacy in society, it is given authority (Høybye-Mortensen 2015). On the other hand, it can be said that if a profession lacks trust and legitimacy, the governing authorities may consider it necessary to regulate it by scrutinizing and gaining more control over its practices, e.g. by imposing new standards and working methods. However, while one argument for digitalizing child welfare, as our analysis shows, is the need to restore trust in the child welfare services, a recent study shows that trust in these services among Norwegian citizens is higher than in many other countries. The same study also reveals that trust has in fact increased in recent years (Skivenes and Benbenishty 2022).

While some studies suggest that digital, standardized systems can restrict room for discretion, assuming that professionals apply standards blindly is a major mistake. Similarly, some studies suggest that ICT and digitalization of work processes can be used as control and management tools (Jorna and Wagenaar 2007; Røhnebæk 2016), but also that frontline workers are not necessarily slaves to digitalized procedures and routines. Studies show how practitioners develop strategies for reducing the negative effects of digital assessment and decision-making tools and apply them in ways they consider best for themselves and their clients (Devlieghere and Roose 2018; Røhnebæk 2016). Røhnebæk (2016) emphasizes how the problem is not ICT systems in isolation, but how these systems are used to manage and standardize work processes that limit frontline workers’ autonomy and discretion.

Our analysis shows contours of the state’s attempts to reduce the scope of street-level bureaucrats by implementing structural measures, such as digital quality and ICT systems, for controlling professional discretion. Irregularities and clients` various life circumstances call for wide discretionary space for meeting clients` needs. Standardized systems restrict professionals’ discretion and ability to meet professional and state ideologies, such as individually tailored services. Nevertheless, standardized systems can never eliminate professional discretion. As Devlieghere and Roose (2018) and Røhnebæk (2016) show, professionals will always seek the best way possible for serving clients` needs. Thus, authorities’ goals of streamlining services and eliminating unwanted variations by digital means, may be difficult, if not impossible, to reach in full.

As our article shows, policy documents are not neutral tools. Rather, they are instruments and carriers of values that affect service provision and ultimately the clients. Certain aspects of ICT such as its ability to measure quality and the decreased scope of street-level bureaucracy call for further research.

Disclosure statement

No potential conflict of interest was reported by the authors.