The coherence of the support instrument mix?
The Smart industry strategy and the area Industry 4.0
In 2016 the Swedish Ministry of Enterprise and Innovation gave the Swedish Agency for Growth Policy Analysis (Growth Analysis) an assignment to deliver an analysis that contributes to the implementation of Sweden’s new industrial strategy entitled Smart Industry. The ministry specified in the appropriation directions that evaluation of the strategy shall begin. It was further highlighted that the evaluation shall, when relevant, take into account the interactions and interdependencies between different support instruments as they affect the extent to which the intended policy outcomes are achieved. The present work was conducted within the scope of this ministerial assignment.
The debate over the new industrial policy has received increased attention in recent years (Warwick, 2013), and policy makers and analysts alike are increasingly focusing on challenges stemming from the policy complexity. This shift in emphasis is exemplified by the uptake of the term “support instrument mix”, which implies a focus on coherence, i.e. the interactions and interdependencies between different support instruments, as they affect the extent to which policy goals are realised (Flanagan, Uyarra, & Laranja, 2011). In addition, the literature highlights that support instruments are not intended to (and cannot) influence the ultimate policy objectives (e.g. structural change) in an immediate sense because these instruments can only influence innovation and industrial development processes. This implies that the ultimate policy objectives must be “translated” into concrete problems that can be influenced directly by support instruments (Borrás & Edquist, 2013).
In this paper we take up the challenge of bringing reflexive learning to the problem of developing and implementing a specific new industrial policy, and we introduce the idea of describing the choice and formulation of its particular “support instrument mix”. We set out the elements we believe are necessary in a model that illustrates the flow from the Swedish Smart Industry strategy, to the instruments deployed in the mix, and finally to the impacts that selected instruments can have; in other words, how the nested relations between policy, instrument mix, and impact align. We suggest an approach to highlight policy coherence that is compatible with a more sophisticated, multi-actor, and dynamic understanding of the processes by which a political strategy emerges, is implemented, and delivers outcomes, i.e. its impact.
Mixes exist on many different unit of analysis. On a higher unit of analysis the literature discusses policy mixes that highlight the interaction between different policy domains such as innovation policy, digitalisation policy, new industrial policy and trade policy. However, this unit of analysis is too wide and too conceptual for our purpose. To our knowledge, there is a gap in the literature when it comes to theories that are used to understand the emergence and implementation of a specific strategy, in our case the Swedish Smart Industry strategy. We aim to further the understanding of how the implementation of the Smart Industry strategy can be improved. Because this requires in-depth empirical evidence of the support instruments in the mix, we delimit this interim report to one of the four areas of the strategy entitled Industry 4.0. The text in the strategy specifies that this area focuses on the digitalisation of industry as is shown in quote below.
“Companies in the Swedish industrial sector are to be leaders of the digital transformation and in exploiting the potential of digitalisation.” (Ministry of Enterprise and Innovation, 2016, p. 30)
Digitalisation is an area of policy making that is characterised by a need for horizontal governance, which implies that the public authorities in charge belong to various levels of authority and policy competences. Effective policies for digital transformation might need to be joined up across a broad set of policy areas, including industrial policy, education policy, research policy, innovation policy, environmental policy, and regional policy. This increases the number of actors and agendas to be co-ordinated in order to achieve coherent policies.
The need for a new industrial strategy for Sweden was part of the prime minister’s statement of the government’s policy on September 15, 2015. The present results highlight that the emergence of the Smart Industry strategy has shown elements of a bottom-up approach. Almost a year of discussions and collaborations, with a large number of triple-helix actors, led up to the launch of the strategy and its related action plan. Engaging the target group and the actors that are supposed to implement the strategy in its emergence is usually a recipe for a success (Vedung, 2016).
Moving on to the Industry 4.0 area of the strategy, we identified the political objectives that should guide the choice of support instruments. The objectives for Industry 4.0 are as follows:
- Stimulating the development, spread and use of the digital technologies that have the greatest potential to lead the industrial sector’s transformation.
- Exploiting the potential of digitalisation broadly, irrespective of industry, company size and geographical location.
- Encouraging new business models and organisational models in order to tap the potential of the new technology.
- Meeting new knowledge requirements that are brought about by digital development.
- Adapting framework conditions and infrastructure to the digital era.
(Ministry of Enterprise and Innovation, 2016, p. 30)
We have conducted a coherence analysis of how the nested relations between policy, instrument mix, and impact align. High coherence indicates that the political objective, in the quotation above, can be implemented. Low coherence indicates that that the instruments in the mix cannot fully implement the objectives. The present results indicate that the political objectives and the instrument mix do not fully align. The strategy was followed by an action plan that specifies the support instruments that make up the mix. It specifies that six new instruments are to be launched in the Industry 4.0 area. Following the flow from the policy objectives, we were somewhat surprised by the choice of instruments. Objective number three and four in the quote above are not followed by instruments in the Industry 4.0 area of the action plan. Interviews, however, showed a more complex picture. Informants described how objective number three, “Encouraging new business models and organisational models in order to tap the potential of the new technology”, has been implemented, but in another area of the strategy entitled Sustainable Production. The fourth objective, Meeting new knowledge requirements that are brought about by digital development, is also described as being implemented in another area of the strategy called Industrial Skills Boost. Because the present analysis is delimited to the Industry 4.0 area of the strategy, and does not cover the other tree areas of the strategy, we asked the following questions. How is the choice of instrument and the instrument’s formulation affected by the fact that the instrument is embedded in the implementation of another area of the strategy? How is the digitalisation aspect taken up in another area of the strategy that has a different focus? Do any conflicts arise when selected support instruments implement several objectives at the same time? Could it be that one policy objective can only be obtained at the expense of another?
Following Borrás and Edquist (2013), we acknowledge that the choice of support instruments constitutes a part of the formulation of the policy and that the instruments themselves form part of the actual implementation of the policy. This double nature of instruments suggests that it is important to look at how they are chosen and formulated. We did this by drawing on insights from text analysis and interviews with the policy makers who wrote the strategy and its related action plan. In addition, we also analysed the programme texts and interviewed all of the programme mangers responsible for all of the instruments in the Industry 4.0 area.
The current empirical evidence illustrates the choice of actors and instruments in the mix. An analysis of the coherence within the instrument mix for Industry 4.0 indicates that one financier instrument focuses more on the leaders and that another financiers instrument reaches out more broadly to include the laggards. Our interpretation is that the Swedish innovation agency Vinnova has the three instruments in the mix that focus more on the first objective in the quotation above, i.e. the leaders. The Swedish Agency for Regional Growth has one instrument that focuses more on the second objective in the quotation above and is formulated to reach out broadly to include the laggards.
One dimension in our coherence analysis was the time perspective. Originally we had hypothesised that results from instruments directed towards leaders could be packaged and disseminated broadly in the instrument used for reaching the laggards. However, the informants described how part of the strategy is expected to be implemented within an electoral period, which is four years in Sweden. The implementation period is therefore perceived as short. Our interpretation is that a short implementation period has resulted in somewhat limited interaction between instruments directed at leaders and the instrument that can reach out more broadly to include the laggards. Instruments directed at leaders are, to a higher degree, designed to generate new knowledge that can be packaged and disseminated. At the same time, an informant described how it can take time for innovation projects to create new knowledge, i.e. results. It is often that results only become available when the project is nearly finished. When results are available, they still need to be packaged, which is an activity that is not always a part of the project itself. Support directed broadly to include laggards, on the other hand, is more designed to disseminate already existing knowledge. When a support instrument is selected and formulated to disseminate already existing knowledge, the following questions become important. Who has generated the knowledge that should be disseminated? Is it the financier or other actors such as intermediaries? Will the knowledge be readily available and packaged when the support instrument is launched? As knowledge is disseminated, what is the receiver’s, i.e. the strategy’s target groups, absorptive capacity?
As a suggestion for future studies, we propose an exploration of pre-existing instruments that the informants have identified as being important for the implementation of the Industry 4.0 area of the strategy. Examples are the strategic innovation programmes, the innovation partnership programmes, regional programmes, and the Industrial Research Forum. Building on existing literature (Flanagan et al., 2011), we acknowledge that the strategy, as with any policy, displays a path dependency. The instruments in the mix are not launched on a tabula rasa, but in a context of pre-existing instrument mixes that have been shaped through successive policy changes. The impact of the Smart Industry strategy therefore also depends on when it is implemented and on the path that has been previously followed.