To Örebro University

Purpose: This paper follows how a Swedish municipality builds their management control infrastructure over 25 years – from the early 1990s to the middle of the 2010s; complemented by a revisit in 2023. The analysis builds on Power’s (2015) framework on “how accounting begins.” As such, it will show the process in which objects such as quality, competence, fairness and economy (1) start their journey as abstract policy objects – promises or dreams, but also as problems – that should be achieved by the municipal organization; (2) begin to take on more concrete forms as they become elaborated, activities are orchestrated and an increasingly stable infrastructure is built around them and (3) achieve multiplicity in ways that make the daily work of practitioners highly complex, leading them to engage in strategizing. This makes accounting continue: once this long and arduous process is worked through, when the infrastructure has stabilized, and the strategizing behavior is learned, this three-step process moves much faster.

Design/methodology/approach: This study is a longitudinal case study over 25 years in one municipal organization: the city of Stockholm – capital of Sweden. We begin in 1991, which is the approximate start of a shift in Stockholm from an old public administration model (Hood, 1991) toward NPM, and we end in 2016, with a post script 2023, when this shift has proceeded into something slightly new and different. Our research engagement with the city of Stockholm has been long-standing: since the early 1990s, we have been conducting five larger research projects on different aspects of management control (for example, the introduction of competition or the decentralization reform) and we have had continuous exchange with city officials and co-workers through workshops and participative observations. For the purpose of this study, we focus on the data from the five research projects and the reading of all budget documents and annual reports from 1991 to 2016.

Findings: The findings extend the framework by Power (2015) in a way that helps us understand how accounting continues. Accounting continues for three interrelated reasons: (1) because of the foundational work that is done over a long time, and according to Power’s (2015) model, where objects such as quality, competence and economy go from existing in abstract programmatic forms to more concrete technological ones around which activities can be orchestrated and where exercises involving learning and collaboration are key. (2) Because such learning and collaboration allow practitioners – professionals and accountants – to engage in strategizing the system in different ways that make it functional. Such strategizing can be argued to be part of the broader infrastructure – not only the “highways” but also the shortcuts and scenic detours are part of it. (3) Because such strategizing makes top managers and politicians think that the system works in rather unproblematic ways – they never see the struggles that go on “the lower levels” of organization. The argument of this paper is that the focus on the outside world and the interest to extend the system by incorporating more and more objects in programmatic forms that we see on the political level in the third period is a sign of further diffusion by accounting: because the system seems to work so well, it can be extended to new parts of the world. And this is how accounting continues.

Research limitations/implications: From the empirical data, we have derived three periods, which can be seen as three phases of building the infrastructure. We see these periods as struggles to gain knowledge about what the organization should provide and an ongoing engagement to build an infrastructure that would support the delivery of the identified objects quality, economy, fairness and competence. These periods could, of course, be divided in different ways. However, as we interpreted the patterns in our data, this dividing of the data seemed reasonable.

Practical implications: In the first chapter of their book, Lapsley and Miller (2024, p. 4) provide a quote from 1983 by an NPM proponent. This proponent depicted the old public administration model, which institutions like, e.g. the NHS, were built upon, as “‘a mobile’: designed to move with every breath of air, but which infact never changes its position and gives no clear indication of direction.” Exactly the same thing can be said about the NPM-related infrastructure we have analyzed. Funck and Karlsson (2023) are obviously right when they argue that proponents of post-NPM movements echo the concerns of NPM proponents “back in the days.” We add to this by arguing that NPM has become what it was supposed to replace: a directionless mobile, maintaining status quo while allowing for strategic behavior (Caffrey et al., 2019). What we learn from NPM’s early days is that no “radical rethinking” (Parker et al., 2023) or “imaginaries” (Funck and Karlsson, 2023) can change this: what is needed, for a real paradigm shift, is the introduction of new problems and new experimentation on the practical level. What remains, both from a practical and academic point of view, would be to discuss what constitutes “enough experimentation”? What can be introduced today that is as disruptive as the introduction of, e.g. tender documents, purchaser/provider models, profit units, etc. during the implementation of NPM? These questions might have to be answered – otherwise accounting will continue to perform as is.

Originality/value: We believe a relevant contribution with the paper is the longitudinal character of the method.