Stefan Henningsson, Jonas Hedman and Bo Andersson
With this paper we inquiry into the possibilities for actors interested in the DI, such as the digital payment infrastructure, to govern and influence its evolution towards specific ends. The inquiry contributes to knowledge about an ongoing transition of e‐Government innovation that moves beyond the scope of digitizing governmental services to challenges of governing and facilitating innovation of public‐private IT infrastructures that are essential in today ' s society. Specifically, through a case study of payments, with particular interest in the Danish digital payment infrastructure we identify and describe the strategies of creating and claiming control over architectural control points. These control points are parts of an DI that have particular strategic importance, for example a monopolized gateway or a de jure mandatory standard for behavior or data interchange.
Although partially successful in the outset to maintain control over the payment DI evolution, the strategy creates substantive barriers for innovation and new actor entry. Ultimately, the lack of innovation leads to that the Danish payment DI and its associated ecosystem is challenged(‘ out innovated’), by a parallel DI supporting payment transactions carried out by a rival organizational ecosystem of which the Danish government has no influence over at all.
2. DIs and architectural control points
The paper builds on the theoretical perspective of DI( Hanseth & Lyytinen, 2010; Tilson et al., 2010; Henningsson & Zinner Henriksen, 2011) and the concept of Architectural Control Points( See, for example, Woodward, 2008). The combination of the two perspectives is an attempt to capture and bridge the interactions between the evolution of DIs and the strategies actors execute to affect their evolution.
In computer science and systems engineering, researchers have long recognized the concept of an architectural control point as a way to identify parts of systems that have particular strategic importance( Woodward, 2008). The concept was developed by the Value Chain Dynamics Working Group at MIT( Trossen and Fine, 2005) in order to understand how commercial benefit is gained from business models emerging in and around the telecommunications industry.
According to Woodard( 2008) architectural control points can be defined as“ system components whose decision rights confer architectural control over other components”( p. 361). Following Woodward( 2008), actors with decision right own the designs that make up the system. More specifically, they hold decision rights that determine, the entities of a system and the relationships between those entities. Consequently, control points can more generally be defined as points at which management can be applied, and any encapsulated functional element of a system can be a control point( Trossen & Fine, 2005). The effect of architectural control points can be small but also powerful, influencing the whole architectural landscape( Elaluf‐Calderwood et al., 2011).
In DI, the presence of architectural control points has been seen as problem. In DIs, the DIs ability to accommodate change in the infrastructure, catering for evolution is considered as key criteria for the DI’ s longterm survival( Tilson et al., 2010). Rodon et al.( 2012) and Hanseth et al.( 2012) report on two cases where the issue of architectural control points was central in hampering DI evolution. According to Rodon et al.( 2012), the Catalan Health Service( CHS) set the foundations for the development of an electronic prescription system in Catalonia by the mid‐2004. In this development project, one of the actors managed through a series of strategic and political maneuvers make modification of the suggested architecture to introduce an additional gateway of which the actor controlled. This design where suboptimal from a technical perspective, and hampered future innovation, but meant that the actor still would keep control of the parts of the healthcare ecosystem. Hanseth et al.( 2012) report on a similar story in the Norwegian health care system, which makes the authors suggest that architecture should not contain any architectural control points that allow individual actors to take control over the whole DI.
Tilson et al.,( 2010) refers to the above discussion as the paradox of control.“ Opposing logics around centralized and distributed control( or individual autonomy) play an equally important role in the evolution of digital infrastructures. This paradox of control brings into consideration the strategic actions of heterogeneous actors and their preferences on modes of control related to change. These considerations shape the services deployed, ownership of data and their definitions, control of critical resources( e. g., APIs), and the appropriation of value.”( Tilson et al., 2010, p. 754)
241