Effective and Democratic Governance as the Condition of Digital Social Innovations in Europe

The world we live in is changing at a fast pace and facing various environmental and social challenges. This applies also to European societies that are facing numerous developmental and institutional challenges. New solutions are needed within organisational and systemic environments, which are oriented towards improving the welfare of society and the wellbeing of each individual. There is a strong demand for social innovations, which imply new concepts and concrete actions in that regard.

In recent decades, the concept of social innovation has gained prominence within social science research. There are various definitions of the concepts, which emphasise the approaches, actions and civil cooperation to provide a sustainable and prosperous society for all. Herein, we take into account the understanding of the concept, which considers social innovation as the process and also as the product. In this regard, the focus is not only on the organisational and social processes but also on their outputs (Baker and Mehmood 2015, 3), bringing forth the solutions to current social demands.

On the one hand, social innovations are supposed to embody the spirit of innovation and entrepreneurship – especially when related to social enterprises. This brings it very close to the modern ideas of development – as progress, based on innovations. On the other hand, it embodies the dimension of ‘the social’, the public good, the ideas of responsibility and solidarity. As such, it is closely related to the ideas of sustainability.

Rapid technological change, as new information and communication technologies play an increasingly important role in all major areas of society, is also affecting the nature of social innovation. These are increasingly based on the use of digital resources to tackle the main challenges and solve the key problems modern societies are facing. In this regard, we can talk about digital social innovations.

Despite the popularity of the concept, the indicators of social innovation are rare. The few existing ones are rather complex and – also due to this fact – only tend to provide the data for a very small number of countries. Relatively comprehensive indicators of social innovation applicable at the macro level include The Social Innovation Index applied by The Economist Intelligence Unit (Economist Intelligence Unit n.d.) and the European Digital Social Innovation Index (EDSII) (Nesta 2019). Although the latter focuses on digital social innovation and is collected at the levels of cities, it is the only one that provides social innovation data for all of the European Union (EU) member states. It is thus also applied and tested in our research.

While considering social innovations as an important aspect of a sustainable society, there is a need to explore the broader context conditioning social innovations to emerge and to flourish in society.

Using the metaphor coined by Giddens (1991), we are in a juggernaut threatening to run out of control. There is a need for steering, which supports and provides new ideas, actions and cooperation to meet all the emerging social needs. Governance plays a crucial role in providing the proper institutional and legislative environment for social innovation to flourish (Galego et al. 2021), engaging collaboration environments of public actors with those from the economy and civil society (Galego et al. 2021; Swyngedouw 2005). The important role is played by participatory and network governance enabling to steer across multi-level scales (Rauschmayer, Paavola, and Wittmer 2009); (Baker and Mehmood 2015, 10). In our research, we understand the concept of active society in terms of quality of governance that combines its effectiveness (as a proxy for steering capacity) with the inclusiveness in terms of democratic political participation, freedom of expression, association and media (as a proxy for polyarchy as ‘real existing democracy as perceived by Robert Dahl [Dahl 1971]). The hypothetical model that serves as a basis of the research is presented in Figure 1.

Figure 1.

The other important factor which we pay attention to in this paper refers to cultural attributes that people share in certain national environments. It has been argued that it is essential to explore shared values among people conditioning similar perceptions, attachments, thinking and behaviour, which potentially lead to social innovation ranging from bottom-up initiatives to general macro-level actions (Gaye Karacay 2021). In this regard, we lean on Hofstede's definition, saying that culture is ‘the collective programming of the mind distinguishing the members of one group or category of people from others’ (‘Hofstede Insights’ n.d.). The Hofstede model of national culture consists of six dimensions. On that basis, we can differentiate between national cultures, which can differently influence social innovations depending on certain cultural orientations and meanings.

In addition to culture, there are also structural settings referring to overall social development, which can influence social innovation to enhance. In this regard, we consider the Human development index (HDI) to play a certain role as well. While usually social innovations are seen as generators for human development (Ziegler, n.d.), we intend to test how the general developmental conditions influence social innovation in the first place.

Accordingly, the purpose of this paper is to determine whether there is a combination of democratic and effective governance conditions in digital social innovation, and the role of human development and culture in that regard.

Based on the theoretical premises and existing empirical research, we thus intend the following:

Present and adapt the indicators of digital social innovation for cross-national comparative analysis of the European countries;

In the fuzzy set analysis, the issue of causality is presented in terms of combinations of potential causes that may or may not lead to a certain effect. If the cause is always present whenever the effect is present, this cause is considered necessary for this effect. If the effect is present whenever the cause is present, the cause is considered sufficient for this effect (C. Ragin 2000; C. C. Ragin 2008). The analysis has been made by applying the truth tables algorithm in fs/QCA 3.1b software (C. C. Ragin and Davey n.d.). Logical operators ‘and’ (*), ‘or’ (+) and ‘not’ (~) can be used between different causes while observing which combinations of causes are necessary and/or sufficient to generate a given effect (C. Ragin 2000; C. C. Ragin 2008). When dealing with crisp sets, the presence of a certain feature is simply seen as members of a given unit in the set of this feature (i.e. having a set membership of 1), while its absence for a certain unit is seen as this unit is out of the set (set membership of 0). When dealing with fuzzy sets, the set membership values closer to 1 are interpreted as the presence of the feature on which the set is based, while values closer to 0 are interpreted as to its absence.

The appropriateness of potential causal combinations – besides their theoretical soundness of course – are assessed in terms of consistency and coverage. Consistency implies the absence of empirical exceptions to the causal rule established by the model. It can be seen as analogous to statistical significance. When the values for consistency are low, especially below 0.8, which is a well-established threshold for raw consistency (C. C. Ragin 2008; Greckhamer et al. 2018, 489), one cannot reject the null hypothesis that there is no causal connection. Coverage, on the other hand, refers to the explanatory strength of the causal model, in terms of how much the effect is explained by the combination of causes provided by the model. Its function is thus somewhat comparable to the determination coefficient in regression analysis (C. C. Ragin 2008).

All of the possible combinations of potential conditions for digital social innovation can be expressed in terms of truth tables. The full complexity of all existing paths towards very high levels of digital social innovations is presented in Table 4. Eight existing causal options have been identified in this regard, which means that they can be seen as alternatives (logical ‘or’) to each other. However, they all share both aspects of governance included in our model, namely inclusiveness (Voice) and effectiveness. This implies that both aspects of governance are necessary for high performance in digital social innovations.

On the other hand, very high levels of digital social innovation are achievable even without very high levels of socio-economic development: while causal paths 1–6 presuppose very high HDI levels as a condition for high social innovativeness, this condition is absent from paths 7 and 8.

The two countries achieving high social innovativeness without very high HDI are Lithuania (path 7) and Portugal (path 8). The former is characterised, in cultural terms, by the lack of power distance and uncertainty avoidance, combined with a long-term orientation. The latter is characterised, on the other hand, by high power distance, uncertainty avoidance, and short-term orientation, combined with the lack of individualism. Both paths share the lack of both indulgence and masculinity. The six paths that involve very high levels of HDI are also quite diverse in cultural terms. Paths 1–3 share high levels of indulgence and the lack of power distance, as well as the lack of uncertainty avoidance. Path 1 includes Denmark, the United Kingdom, the Netherlands and Sweden, combined with high individualism and the lack of uncertainty avoidance. In Denmark (path 2, which also includes Finland and Sweden), this is also combined with low masculinity and low long-term orientation. On the other hand, Germany and Austria are examples of path 3, characterised by high long-term orientation and masculinity.

Paths 4–6 all involve high-power distance and uncertainty avoidance. France and Belgium (path 4) combine that with individualism, long-term orientation, and indulgence. Paths 5 (Slovenia) and 6 (Czechia) are more specific. While they both share the lack of individualism, Slovenia has lower masculinity and more short-term orientation when compared with Czechia.

While this diversity of paths may be quite interesting, it also indicates the major drawbacks of this complex explanatory model. We can note eight different paths, but most of them consist of a single case. This may imply some particular (national) specifics that could hardly be generalised in any persuasive way. In addition, the small size of our sample combined with a great number of sets (i.e. two for governance, six for culture and one for development) inevitably produces a high number of empirically non-existent combinations that cannot be tested. In the language of fuzzy sets analysis, such cases are called counterfactuals, as they cannot be found among our actual cases. As such, combinations would be logically possible – at least based on Hofstede's theory, which sees the cultural dimensions as mutually independent and thus potentially compatible with each other and with different levels of socio-economic development, one cannot exclude the possibilities that also several other combinations could lead to high levels of digital social innovation.

In addition, the numerical parameters of the complex model may also question it. While its overall consistency of 0.947 is beyond any doubt, its overall coverage of 0.700 is far from impressive. While the model is highly consistent, its overall explanatory power is rather weak.

Consequently, the claim that high levels of particular cultural features and/or very high levels of socio-economic development are necessary for high levels of digital social innovation can hardly be supported. One can note the exceptions (Portugal and Lithuania) generating high levels of social innovations despite – in European terms – relatively low levels of socio-economic development. This proves that very high levels of development are not necessary in this regard. The same may be claimed about cultural features, as several culturally diverse paths produce similar levels of digital social innovations.

On the other hand, the two conditions related to governance seem to be necessary, since they appear in all eight paths identified by our complex model.

All of this calls for a more parsimonious explanation that may drop the unnecessary elements from the explanatory model. Such a model is presented in Table 6. It is a combination of inclusive (Voice) and effective (Effect) governance with both conditions being necessary and thus allowing no alternative paths. When compared with the complex model in Table 4, it provides a significant increase in coverage, namely from 0.700 to the quite persuasive 0.879, while its consistency of 0.920 remains fully satisfactory (i.e. only 0.027 lower than for the much more complex model presented above). The parsimonious solution dropping the aspects of culture and socio-economic development as rather irrelevant is thus much sounder than the complex one.

Even when only observing the two aspects of governance, we should note that our population of countries includes no countries with high effectiveness and no high inclusiveness. Whether the non-existence of the high effectiveness–low inclusiveness combination among the selected European countries is just a coincidence, is beyond the scope of this research. It implies, however, that the causal effects of effectiveness cannot be tested separately but only combined with inclusiveness – as nobody can check the effect of a nonexisting phenomenon.

The results of causal Model 2 are presented in Figure 2. We can observe a high level of consistency between lower governance results and lower social innovation on the one hand and the high governance results and high social innovation performance on the other hand. Estonia is the only major exception in this regard – with comparatively good results in terms of effectiveness/inclusion but weak digital social innovation.

Figure 2.

As inclusiveness and effectiveness of governance are necessary and sufficient for digital social innovation (with a high level of consistency), there can be no cases with inclusive and/or effective governance but with no high digital social innovation. The final parsimonious model that drops the conditioning by culture and development that could not be confirmed is summarised in Figure 3.

Figure 3.

European societies have been facing several societal challenges, the success of which will have a major impact on the position of the ‘old continent’ in a global context. The response of political institutions, both at the national and EU levels, to the crises of this millennium – from the financial to the migrant and the current epidemiological crisis – has been far from optimal. Inefficiency and irresponsibility of traditional political parties and their lack of leadership lead to poor performance in tackling these problems, contributing to a decrease of political trust in many European countries. There is a need for improving the quality of governance to maintain a high level of systemic competitiveness as well as a high level of well-being and social cohesion.

Governance is strongly linked with social innovation. As shown by our explanatory model, dimensions of governance, inclusiveness and effectiveness are necessary and sufficient causes of digital social innovation to take place. This means that existence of channels for widespread participation of the people, including vulnerable and marginalised groups, in political life, on a basis of their autonomous choice, as well as effective and transparent conduct of political institutions, staffed by competent people with a high level of professional and personal integrity, lead to a boom of creativity and utilisation of human capital in all key areas of society. This brings more optimal solutions for resolving political, economic, social and technological problems that burden contemporary societies. The tight relationship between governance and social innovations is particularly relevant regarding the pursuit of sustainability in terms of the development and implementation of new innovative policy ideas to tackle key environmental challenges (Baker and Mehmood 2015; Parra 2013).

When we talk about the role of democratic governance and its societal effects, emphasis is needed on the existence of a balance between its dimensions. Neither participation alone nor sheer effectiveness is sufficient. They only contribute to the development of innovation through mutual interactions.

It has turned out that neither cultural features nor the level of socio-economic development can be proven as relevant as separate factors in stimulating digital social innovations. A note of caution is needed in this regard. First, it should be stressed that this conclusion may only be drawn for the European or even more specifically EU context, where the disparities either in terms of culture or in terms of socioeconomic development are not much less significant than at the global level. We are dealing with countries that are culturally still relatively similar and relatively well-developed when compared with most other parts of the world. According to the UNDP HDI (United Nations Development Programme n.d.), all countries in our analysis are ‘highly developed’ despite their differences. The lack of explanatory potential of the socio-economic development and culture can thus be related to these facts. And second, the lack of direct conditioning of digital social innovations by culture and development does not imply that there are no background effects of these factors on the inclusiveness and effectiveness of governance. Our additional fuzzy sets analyses indicate that culture (but not socio-economic development) may be necessary and sufficient condition for effective and inclusive governance of the European countries.

Consequently, it might be argued – though with some caution – that high inclusiveness and effectiveness as components of quality of governance presuppose particular cultural characteristics since these factors strongly overlap. However, again we have to state that the importance of particular cultural characteristics as detected in our analysis holds for European societies and thus one shall avoid generalisation since it is possible that in non-European environments, different cultural combinations are more relevant.

At least in the EU context, our results provide some reasons for optimism. Among the EU member states, the highest levels of socio-economic development are not necessary for impressive performance in terms of digital social innovation. Even the comparatively less developed EU countries and EU countries with a significant cultural variety are capable of high levels of social innovation – as a major source of sustainable development.