Stimulating civic behavior? The paradoxes of incentivising self-organization

ABSTRACT

Crowding theory has highlighted the unintended consequences that well-meant financial incentives can have on behaviour. Using field research, this study applies crowding theory to self-organising citizens’ initiatives in the Netherlands. Citizens’ initiatives create different kinds of goods and services and are sometimes supported by governmental ‘citizens’ initiatives stimulation funds’. We found that (1) the crowding effect is multifaceted; (2) the psychological processes and mechanisms of crowding in and out are subtle and can happen simultaneously; (3) in some instances, stimulation funds seem to have affected the preferences of the initiatives, such as expectations about the funds’ presence and procedures; and, (4) stimulation funds are well-suited to instrumentally support the initiatives in realising their plans and ambitions, but less suited in fostering civic behaviour or spontaneous self-organisation. Our research raises important considerations about the possibilities and constraints for stimulating civic behaviour through stimulating citizens’ initiatives.

KEYWORDS:

• Citizens’ initiatives
• civic behavior
• civil society
• crowding out effects
• crowding theory
• self-organization

Introduction

Citizens taking initiative in organising public goods or services are increasingly seen as indispensable partners to realise collective goals (Healey 2015; Van de Wijdeven 2012; Igalla, Edelenbos, and van Meerkerk 2019). Accordingly, governments encourage citizens to participate in public affairs through various means (Brandsen, Trommel, and Verschuere 2017; Van Buuren 2017). But encouraging citizens may also have adverse effects. Various scholars have argued that financial support as compensation for doing something out of a civic spirit can, in fact, undermine that spirit (Frey and Oberholzer-Gee 1997; Alford 2002; Bowles 2016). For instance, experiments show that fines meant to incentivise parents to bring in their children to the day-care on time can actually contribute to making them come in late (Gneezy and Rustichini 2000). The common presumption is that the shortcomings of individuals should be neutralised by interventions from the state (Ostrom 2000, 7). However, this presumption calls for rules and incentives that can invoke exactly the type of behaviour that they are intended to prevent.

As stated above, governments aim to support citizens’ initiatives (Brandsen, Trommel, and Verschuere 2017, 677). Sometimes this support is deemed necessary because it is said to be the key ingredient in fostering the right climate for citizenship to evolve (Tonkens 2006). Governments are also expected to ensure that participants in citizens’ initiatives will not be suckered – i.e., taken advantage of – by free riders (Bowles 2008). This article focuses on the question of what happens if a government incentivises the self-organisation of citizens’ initiatives. An external incentive, for example, a fine or a subsidy, may crowd out intrinsic motivations and instead induce extrinsic motivations (Frey and Jegen 2001). When intrinsic motivation is crowded out, self-esteem and self-determination are impaired. This means that ultimately initiators are discouraged and no longer motivated to collectively organise. To understand how crowding out works in relation to citizens’ initiatives, this article answers the following research question: To what extent and under what conditions do financial incentives by local governments aimed at citizens’ initiatives crowd out motivations of the initiators of these citizens’ initiatives?

Supporting citizens’ initiatives can be considered as a balancing act for local governments, where wrong decisions about incentives can be unintentionally costly in the long term (Vollan 2008, 563; Bowles 2016; Brandsen, Trommel, and Verschuere 2017). Thus, by focusing on this balancing act, this article contributes to our understanding of effective governmental facilitation by applying crowding theory to the phenomenon of citizens’ initiatives. Second, as crowding effects are usually studied using experiments (Bowles and Polania-Reyes 2012), we contribute to this body of knowledge by applying crowding theory in field settings to see how it works outside the labs. Finally, crowding theory is mostly known from and studied within economics or psychology with a strong focus on financial components. Crowding theory clearly states that any intervention may spark its crowding out effects (Frey and Jegen 2001; Cardenas, Stranlund, and Willis 2000; Bowles 2016). We make a broader application of crowding theory by focussing on local funding schemes, a policy instrument that consists of a mixture of financial incentives, rules, and procedures.

In the remaining part of this article, we first set out some important theoretical building blocks about citizens’ initiatives, crowding theory, and motivation. Next, we discuss the methods of data collection and analysis. We then present the results of our research. Finally, we draw conclusions about the crowding effect of stimulating citizens’ initiatives and discuss the findings from a theoretical and empirical perspective.

Motivating active citizens

Governments view active citizens as important because active citizenship is seen as beneficial to both the individual and society (Tonkens 2006, 9; Van Dam, Duineveld, and During 2015, 165 and 175). Just like the concept of the Big Society in the UK (see Healey 2015), in the Netherlands, ‘the Participation Society’ was coined to describe and prescribe the active relationship between citizens and governments. As such, the current emergence of citizens’ initiatives fits well in the government’s view on how citizens should behave in public life (The Ministry of General Affairs 2007, in Boonstra and Boelens 2011, 101). An illustration of the various efforts that the Dutch government has unleashed to perpetuate the movement of public participation and citizens’ initiatives are ‘the Do-democracy’, ‘Into action with citizens’ (Van Dam, Duineveld, and During 2015, 164), or more recent examples such as ‘Democracy in Action’ (Ministerie van Binnenlandse Zaken en Koninkrijkrelaties 2018).¹ Citizens’ initiatives thus fit well in the government’s agenda.

Because of this agenda, governments also have an interest in supporting citizen’s initiatives. Governments often give external incentives to reinforce motivation or to show what it sees as desirable (e.g., Vollan 2008, 560). Financial or organisational support by governments is seen as an important asset in making citizens’ initiatives function properly (Nederhand, Bekkers and Voorberg 2016, 1079). Tonkens (2006, 15) argues that ‘active citizenship only comes about in an institutional stimulating context: viz, in interaction with, and supported and stimulated by active governments’.

Especially local governments are fond of citizens’ initiatives and at the same time have a tendency to steer and control them with help from the rules that accompany their support (Streeck and Schmitter 1985, 122). Boonstra and Boelens (2011, 107) summarise the difficulty of the relationship between governments and citizens’ initiatives. ‘Governments ask for citizen involvement and shared responsibility, but (…) hold on to instruments that keep them in central positions, in a society that is no longer governable from that kind of one-dimensional perspective’. Van Straalen, Witte, and Buitelaar (2017) provide evidence for the difficult relationship between a ‘controlling’ government and self-organising citizens.

In summary, governments often envision the ‘right behaviour’ of citizens and aim to incentivise that behaviour with subsidies or other resources. We argue that supporting citizens’ initiatives conforms to the governmental vision of self-reliance, self-sufficiency, and civic behaviour by citizens. Support may very well be beneficial to citizens’ initiatives, but Brandsen, Trommel, and Verschuere (2017, 676) warn that ‘however commendable many initiatives may be, there is the risk that the desire on the part of governments for their citizens to participate and self-organise may lead the state to take over such initiatives’. This leads to a manufactured civil society where no spontaneous self-organisation takes place as opposed to what takes place in ‘true’ civil society. We further explore the drawbacks of incentivising citizens’ initiatives through crowding theory.

Crowding theory

Crowding theory is mostly known for the crowding out effect: the ‘theoretical possibility that intrinsic motivation may be negatively affected when a previously non-monetary relationship is transformed into an explicitly monetary one’ (Frey and Jegen 2001, 590). Crowding theory acknowledges the fact that both crowding in and crowding out can occur (Frey 1997, 1044), we therefore use the general term crowding theory, or we refer to the specific direction of crowding in or out. Crowding theory is known to originate from Titmuss (1970) ‘The Gift Relationship’. He wondered how blood donations differ in quality and quantity based on how the donation system was organised: public and based on volunteers or private and based on payments. Though, at that time, without an explanation why altruistic motives supposedly outclass self-interested motives in blood donations, social psychologists coined ‘the hidden costs of reward’ (Frey 1997, 1044). The hidden cost of reward is the decreasing or diminishing intrinsic incentive to act whenever an external intervention comes into play. Crowding out effects have been demonstrated by experiments in a NIMBY setting (Frey and Oberholzer-Gee 1997). Residents were asked if they welcomed a nuclear waste repository in their community. Almost half of the residents agreed out of civic spirit. When a compensation was introduced – a supposedly supporting incentive – acceptance rates sunk below 25%. Intrinsic motivation to do something for the public good was crowded out by extrinsic monetary incentives. Another well-known experiment is by Gneezy and Rustichini (2000) on day-cares: to reduce the number of parents who collected their children late, a fine was introduced. After the implementation of the fine, the number of late-coming parents increased. The parents saw the fine as a price to pay as compensation for being late.

Crowding theory basically views intervention as followed: all interventions originating from outside the person under consideration (i.e., both positive monetary rewards and regulations accompanied by negative sanctions) may crowd out or crowd in intrinsic motivation or leave it unaffected (Frey 1997, 1045; Frey and Jegen 2001, 592)

While crowding theory is mostly known for crowding out, because intervention is perceived as controlling (Frey and Jegen 2001, 595), the theory clearly states that interventions can also crowd in, strengthening the desired effect at hand. This often happens when an intervention is perceived as supportive. Crowding not only occurs by imposing a monetary intervention; other measures – for example supervision (Bowles 1998, 91) – may trigger effects as well (Frey 1997, 1048; Cardenas and Ostrom 2004, 312; Ryan and Deci 2000a, 59). For crowding out, the following psychological processes happen because of an intervention (Frey and Jegen 2001, 592):

• Impaired self-determination. When an intervention is perceived as reducing self-determination, extrinsic control replaces intrinsic motivation. A person feels no longer responsible and feels no longer any autonomy. Intrinsic motivation is redundant next to a steering and controlling intervention.

• Impaired self-esteem. If an intervention signals that intrinsic motivation is not valued, e.g., through a monetary compensation, the individual is deprived from showing and fulfiling his or her own interest and involvement. Reduced effort is the result.

Why then, do interventions reduce self-determination or self-esteem? Samuel Bowles (2016) argues that there are four ‘mechanisms’ that ‘trigger’ and constitute crowding out (Bowles and Polania-Reyes 2012). These mechanisms should not be interpreted as mechanically determined/mechanistic, rather these mechanisms are mental/psychological in nature. They produce effects. The first mechanism is ‘bad news’: the incentive signals the principal’s (in this case, the government’s) negative beliefs and opinions about the agent (a citizens’ initiative). For example, strict rules signal that the principal thinks that the agent is untrustworthy, lazy, and self-interested. According to Ostrom (2005, 267) two messages are currently told through controlling policies: first, that citizens are selfish and therefore need regulation. Secondly, that citizens are uncapable of solving complex problems and professionals should solve these. Both messages are illustrations of bad news. Citizens’ initiatives stimulation funds (CIS-funds from here on) with strict rules or conditions could, in fact, trigger bad news.

The second mechanism is ‘moral disengagement’: the incentive creates a new, more instrumental perception of the situation that the agent is in. In the day-care example (Gneezy and Rustichini 2000), collecting one’s children late is put in a market-like situation where it receives a price, resulting in market-like considerations by the parents. The parents concluded that the fine is worth paying. The fine invoked a moral disengagement and created a rational cost-benefit analysis. When initiatives are only active because of the governments’ subsidy and for no other reason, initiators can become morally disengaged

The third mechanism is ‘control aversion’ and directly affects self-determination: an incentive that is perceived as a means to gain or increase control, restricts autonomy, and therefore negatively affects the agent’s sense of self-determination. When the application or accounting procedure for the subsidy is overly strict, initiators might experience control aversion.

The last mechanism is ‘endogenous preference updating’. This is less like a trigger compared to the previous three mechanisms. As opposed to the other mechanisms, which occur or work within the ‘action situation’ (where people make choices) and are seen as exogenous, preference updating is endogenous and does not influence the action situation like the other interventions do. Bowles and Polonia-Reyes (ibid., 374) explain it as follows: ‘The type and extent of a society’s use of economic incentives also may affect the process of preference-updating by which individuals acquire new tastes or social norms that will persist over long periods’. In other words, because of the repeated use of incentives, individuals learn other preferences, preferences which may respond well to incentives. This, ultimately shapes the action situation more indirectly.

Motivation

The one question that we are left with before we can synthesise crowding theory and citizen’s initiatives is: to what extent are citizen’s initiatives based on intrinsic motivation? ‘One is said to be intrinsically motivated to perform an activity when one receives no apparent reward except the activity itself’ (Frey and Jegen 2001, 591). Extrinsic motivation is usually associated with doing something for rewards or evading punishment. Ryan and Deci (2000a; 2000b) argue that extrinsic motivation is essentially based on how people are motivated to act, ‘do something’ or acquire certain ends aside from the activity itself. They also argue that extrinsic motivation is more like a continuum based on how self-determined an actor is. Individuals who act because they find some action important or because it reflects their values are very self-determined but ultimately extrinsically motivated. These individuals are acting not because it is ‘fun’ in itself, but for other end-oriented reasons. Ryan and Deci (2000a, 2000b) distinguish between four types of extrinsic motivation:

• external regulation (least self-determined) is about acting because someone demands it;

• introjected regulation is about acting to avoid a sense of guilt;

• identified regulation is about personal importance; and

• integrated regulation is the most self-determined form of extrinsic motivation, but it is still about achieving ends that the individual deems valuable.

Triggers that motivate people to do ‘something’ or solve a problem are important for citizens’ initiatives to start acting (Hurenkamp, Tonkens, and Duyvendak 2006, 20; Nederhand, Bekkers, and Voorberg 2016). As such, it is unlikely that this motivation is purely intrinsic. Citizens have ends, goals, and ambitions in mind that they want to pursue. Initiating an action is then likely a more self-determined type of extrinsic motivation. Hurenkamp, Tonkens, and Duyvendak (2006) show that many people self-organise for certain goals or reasons: because of a personal experience, an experience by others, or because of a social happening. All these reasons for self-organisation are not associated with purely intrinsic motivation. Neither is it the type of extrinsic motivation that is usually associated with punishments or reward. Citizens’ initiatives are not developed because someone forces citizens to do so or through rewards and punishments. It is most likely that citizens have their own reasons to act. In other words, they are extrinsically motivated by a strong sense of self-determination.

To summarise, citizens’ initiatives exist because their initiators and participants are either extrinsically motivated with a high level of self-determination or intrinsically motivated. Therefore, their motivation can be crowded out. But motivation can also be crowded in. Based on the literature, we present four factors that affect how a crowding effect arises among citizens’ initiatives when CIS-funds get involved. We inquire into the following issues:

1. What type of motivation is at play;

2. Whether the fund is perceived as supportive or controlling;.

3. Whether self-esteem or self-determination has been affected;

4. What mechanisms are involved and how they are involved.

This leads to the following analytical framework with four types of variables and their possible values (Table 1).

Table 1. Variable and possible values.

Variables	Possible values
1. Motivation	Extrinsically (with lower/higher degrees of self-determination) Intrinsically
2. Perception	Supportive/controlling
3. Processes	Self-determination: to a lesser/greater extent
	Self-esteem: to a lesser/greater extent
4. Mechanisms	Bad/good news
	Moral engagement/disengagement
	(An absence of) control aversion
	(Any form of) preference updating
5. Results	Is there a crowding in or out effect?

Methods and design

Data collection for this article has taken place through eight case studies of municipal incentive schemes for citizen’s participation with eight initiatives that receive subsidies from these schemes. Case studies are suited for our research purposes as we are testing if and exploring why citizens’ initiatives are prone to crowding based on the analytical framework. In that sense, our study follows a deductive logic. A case study design matches the multi-causality and probable fine-grained appearance of crowding effects in real life situations.

Several municipalities in the Netherlands have CIS-funds to invite, entice, or support citizens’ initiatives. We searched for funds that explicitly target citizens’ initiatives as appears from their name: ‘Initiative Fund’. This name distinguishes them from other municipal schemes that provide regular subsidies to associations. We found a dozen and included eight. Out of the four cases that were excluded, one municipal fund had no interest in participating in the research, two funds were unable to establish contacts with users of the funds and in the final one there were no initiatives that draw from the fund yet. The selected CIS-funds are all relatively young, maximally four years old because we assumed that older CIS-funds have experienced learning effects (see for example Van Straalen, Witte, and Buitelaar 2017). The CIS-funds are all rather similar in terms of their purposes, rules, and additional conditions. The amount that one can maximally apply for and the total available money differ. An overview of the CIS-funds and their characteristics is available in the online supplemental material.

First, we conducted short unstructured interviews of approximately 30 minutes with a civil servant or volunteer of the fund and selected the initiatives in deliberation with the drawees. Selection criteria that we set and met were similarity and representativity. Similarity is important for our purpose to see how crowding theory applies to citizens’ initiatives (Gerring 2006, 131–139) as most-similar selection is very well suited to confirm if crowding theory applies to citizens initiatives and explore how it works. Representativity is important for exploration and prevents that our cases are outliers. The selected initiatives were similar and representative in terms of the domains they were active in (Hurenkamp, Tonkens, and Duyvendak 2006).

Second, we visited the initiative and interviewed one applicant from an initiative per fund about their activities and experiences with the fund over the course of the second half in 2019 and the beginning of 2020. Citizens’ initiatives are often dependent on a small core group of active citizens and as types of motivation are important in this study, we interviewed the individual that was considered the driving force behind the initiative. We conducted semi-structured interviews (Gabrielian, Yang, and Spice 2007, 156) of one hour with the initiators about the daily situations of and in their initiative. We used an interview protocol with the following themes: (1) the general characteristics of the initiative; (2) the goals, ambitions and motivations of the initiators; (3) how they perceive the fund; (4) what the CIS-funds does for/to them; and (5) how initiators perceive the subsidy, the rules, the procedures and the involved civil servants. Interviews were recorded, reviewed and summarised. We analysed the interviews by coding all relevant statements about the variables in the analytical framework per case and we interpreted the data both ways. For example, CIS-funds could signal bad and good news. Descriptions of the cases were constructed based on the analysis with: (1) a short general description of the case and (2) perceptions of initiators and comprehensive storylines (Blatter and Haverland 2012, 111 and 115) with detailed attention for the ‘little things’ about the four mechanisms that make the crowding effect. The descriptions were reduced to the shorter analysis (in the next section) per variable to present a comprehensible overview of our findings (see the online supplemental material for the extended descriptions).

Results

First, we provide short descriptions of the cases followed by an analysis of each of the four categories from the analytical framework, and then we finish with a summary of our findings (Table 2).

Table 2. Cases and descriptions.

No.	Description
1	This initiative organises all kinds of activities focused on increasing social cohesion and unity in the neighbourhood. Examples are fishing together, bus trips, ‘neighbour-day’, maintaining gardens and the particular activity supported by the fund of renovating a small garden
2	The activity is for improving and maintaining a playground in a specific block of a neighbourhood. The active neighbours in this initiative consist of roughly seven families and their activities regarding the playground was a one-off occasion.
3	The orchard is located in a green space between two rows of houses in a relatively new neighbourhood. The initiator and other citizens were not satisfied with the green space, and they made a new plan for the green space, turning it an orchard. They adopted some trees that the municipality wanted to get rid of and organise activities in the orchard.
4	The food garden is located in one of the Dutch COA’s (Centraal Orgaan opvang Asielzoekers, roughly translated as ‘central agency for the reception of asylum Seekers’. The initiators organise the garden in collaboration with the refugees.
5	Inspired by the counting of birds in the area and a recent lecture on butterflies, the initiators created a coalition with a local (retired) farmer who would make his land available, a school, and the municipality to create the butterfly garden.
6	This initiative involves redecorating structures. One was ‘the cage’ – a local gas distribution point – and the other was ‘the shack’ – a local hanging out spot for the youth – both in decayed state. The initiator, who is a teacher at a secondary school, thought of a project to let students aid in the renovation and decoration of the structures. Students, kids from the local primary school, the municipality, a cultural organisation (with a graffiti artist), and the gas company made the structures more attractive
7	This initiative is a group of women who aimed to create a natural playground in their village for the kids. The initiators turned first to the developer of some of the lots in the village but with no results, and they then went on to the municipality and other partners to realise their ambitions.
8	This initiative is about a group of kids working weekly on robotics and who train/prepare for robot competitions. The initiative in its current form ‘just came about’ because the parents got involved in the kids’ activities.

In the following sections, the findings of all the cases collectively are presented variable by variable. We conclude this section with a final verdict based on our findings.

Type of motivation

In none of the studied initiatives have pure forms of intrinsic motivations been found. Most of the initiators were motivated because they wanted to achieve something. They had concrete ends and goals, mostly in their direct neighbourhood, which they wanted to acquire. Initiators acted because they found it personally important and valuable to achieve those results: parents wanted to facilitate activities for their kids so that they could work on robotics, and locals wanted to renovate a garden, improve or create a playground, or create a beautiful orchard. Based on Ryan and Deci’s typology, identified regulation (high self-determination) was the most dominant type of motivation.

In three of the cases, integrated regulation (most self-determination) was at play. In the food garden, for example, the initiators were driven by deep values to change the welfare state, and one of them had been working on that for decades. At the same time, they were practically oriented towards their goals: they wanted their project to succeed and needed funds and partners to accomplish this. Also, in the butterfly garden and the redecoration/renovation cases, more self-determined forms of motivation were visible. The initiators of the butterfly garden were enthusiastic and involved with nature and since they were young, they wanted to pass on this enthusiasm and involvement to local school kids. In addition, they also want to help improve the current state of nature regarding bees and butterflies. In the redecoration case, the initiator was also an artist looking for inspiration.

Perceived support or control

Most of the citizens’ initiatives perceived the CIS-funds – in general (we cover the application procedures later in the section about the mechanisms) – as supportive and not as controlling. In half of the cases, the initiators were steered to the fund by civil servants, showing that CIS-funds are actually looking to support active citizens rather than tightly controlling and suppressing applications. This attitude of support is affirmed by the initiators from the food garden:

They were not looking for reasons to say ‘no’, they were looking for reasons to say ‘yes’. And they also said: if you need more money later, please come back to us. After the report, which was very short, they also asked if they could have pictures. So they are very supportive.

This perception of support is mainly the result of civil servants often being involved and helpful in the projects. The initiators of maintaining the playground were fond of the civil servants who were enthusiastic; the initiators of the butterfly garden mentioned that a civil servant helped with the application; and the initiator from the natural playground was aided by civil servants in making their project (plan) better. Other initiatives felt supported (and esteemed) because of a visit by a local council member to view the results. Consequently, in the end, none of the CIS-funds were seen as controlling by the initiators. At the same time, the initiators noted that they had other experiences with other non-governmental funds who had more controlling conditions or were more distant.

In two of the cases, renovating the bedding and the orchard, the CIS-funds were neither perceived as controlling nor as supportive. Instead, the initiators saw the CIS-funds and their accompanying rules as ‘part of the deal’ (we will get back to this later in the section on mechanisms).

Influenced self-esteem or determination

We did not find any instances of impaired self-esteem. The incentive did not devalue the motivations of initiators. Rather, in more than half of the initiatives (renovating the bedding, the food garden, the butterfly garden, the redecoration case, and creating the natural playground), the self-esteem of initiators increased because of involved, appreciative, and helpful civil servants or a local councillor (they felt respected for the same reasons as they felt supported).

Self-determination was not impaired; instead, we found indications that self-determination was actually ‘nurtured’ by the fund. In three cases, maintaining the playground, the redecoration case, and the ‘kids and robotics’ case, the CIS-funds were not necessary for the projects because, according to the initiators, they were sure that they would have made it work either way. In those cases, however, the fund did offer the possibility to create better results. In the other five cases, the CIS-funds were indispensable for the initiatives to be realised. At first sight, the funding seems somewhat transactional, but the intervention seemed to do the opposite of replacing self-determined motivation: it facilitated it. Because of the fund, initiators were able to pursue their (non-intrinsic) motivations and create the results they aimed for: ‘If we didn’t get the fund, we wouldn’t have made it financially’ (renovating a garden); or ‘You want the project to succeed. And I know that if we didn’t get money elsewhere, we would have a problem’ (the food garden); or ‘If there are no resources, people often don’t even start. A playground, for example, people are not going to pay for that. You need the fund’ (the butterfly garden). To cite Frey and Jegen (2001, 595), initiators ‘feel that they are given more freedom to act’.

Mechanisms

We distinguished four mechanisms that can provoke crowding effects. First, there were no instances of the CIS-funds signalling some kind of bad news in the cases that we examined. Most often, initiators did not perceive any signal at all or they perceived a positive signal. The good news was, in the cases of maintaining a playground and the butterfly garden, that governments want locals to take the initiative. Other examples of this positive news are the following statements ‘It’s some form of appraisal for initiatives’ (redecoration case) and the initiators of the food garden said that ‘Money from the CIS-fund, which is run by a local organisation, has the benefit of the local approval’ (as opposed to a general fund). The fund did not signal abstract good news to them, but rather they saw it as some form of good news legitimised by the neighbourhood community.

Secondly, we did not find any signs of control aversion. The conditions for the application and the process of applying were seen as accessible and logical, rather than Kafkaesque and full of bureaucratic red tape. Furthermore, accounting for the received money was also generally seen as fair and accessible. Some initiatives were allowed to show photos of their results as accounting for the fund, while others needed to present a balance sheet and bills. None of the present conditions nor administrative processes invoked control aversion.

Thirdly, we did not find any moral engagement; however, we did find multiple – but only slight – instances of moral disengagement. As is illustrated through quotes in the section on self-esteem, most citizens’ initiatives need the CIS-funds or otherwise they would not exist. Some initiators (of the orchard, the food garden and the natural playground) were involved in applying for different funds or looking around for other funds. There were also instances where an initiator saw the CIS-funds as measures of ‘outsourcing tasks’ by the municipality. Other initiators noticed the tension between moral engagement and financial needs/support. The initiators of the butterfly garden discussed with one another how often money is necessary to realise projects. And how without funds, it would likely be difficult to start these civil activities. But in general, for these initiatives, their situation requires action which is framed in such a way that stresses the need for money.

Lastly and most surprisingly, we have seen multiple references that hint at some sort of conformation to updated preferences in all of the cases: initiators expect some jar of money accompanied by a set of rules or conditions because it is ‘part of the system’. In all but one case (robotics for kids), the initiators provided reflections that resemble conformation to such preferences. The initiator that renovated a garden said: “But it is also part of ‘how we do things,’ the fund is part of the whole, it has to be there”. Another initiator (maintaining a playground) expects to account for the subsidy: ‘they want to know where their money goes. Something like that is part of the deal’. Also, in the cases of the food garden, the butterfly garden and the redecoration case, initiators mention it is good to account for the support given by the fund. It is ‘taxpayers’ or public money, after all’. Citizens’ initiatives start, expecting that funds are available to make things possible. The presence of CIS-funds is seen as ‘natural’ and initiators anticipate on its availability. The way in which respondents talked about the CIS-funds reveals that they see their initiative as part of the rules of conduct: governments that want to welcome initiatives have to make financial support available.

Crowding effects: the final scores

Through our findings, we conclude that crowding effects are highly multifaceted. First, in the cases that we have analysed there are no strong crowding out or crowding in effects. The closest finding representing a full crowding effect – people quitting or starting – was one CIS-fund that inspired citizens to discuss among one another whether to apply for the funding.

Secondly, many of the CIS-funds are actually seen as supportive rather than controlling, but support was mostly perceived by the involvement of people of the governments showing appreciation. Self-esteem and self-determination did not seem to be impaired.

Finally, we saw that many CIS-funds signal good news to the initiatives and do not invoke control aversion by means of their conditions for application and procedures. These findings are all pertinent in explaining why we did not find any crowding out effects. Even though the CIS-funds showed more positive/supporting interpretations of the four mechanisms (good instead of bad news), and considering that self-determination and self-esteem seemed to be facilitated, we surprisingly also did not see any definitive crowding in effect. To explain why, we turn to the lack of moral engagement. In many of the cases there was a slight presence of moral disengagement because financial resources were necessary to realise the initiators’ goals. External financial support is often indispensable and acquiring it requires considerable attention from the initiators. Without this money, some initiatives just do not take flight. Most initiators know this and seem to act accordingly because their goal is important to them. They were motivated – mostly through identified regulation (high self-determination) – to realise their goals. Identified regulation should be receptive to incentives that support realising those goals, provided that they are not perceived as controlling. The CIS-funds did not undermine the motivations and impair self-determination. But the CIS-funds also did not trigger or increase motivation and did not morally engage.

Conclusion and discussion

The central question of this article was: To what extent and under what conditions do financial incentives by local governments aimed at citizens’ initiatives crowd out motivations of the initiators of these citizens’ initiatives? By dissecting the crowding effect, we were able to show how multifaceted this effect is. We did not find the expected controlling conditions nor an extensive crowding out effect as predicted by theory. However, it is striking that financial interventions did not crowd in either. Initiators felt respected and supported by appreciative civil servants and less so by subsidies. The CIS-funds that we investigated left considerable autonomy to the initiators and help them realise their goals; the CIS-funds seemed to enhance self-determination to some extent. The CIS-funds also signalled good news and did not trigger control aversion. Simultaneously, we found that CIS-funds seemed to invoke slight moral disengagement and we found hints of updated preferences. The paradox of incentivising self-organisation appears to be absent: governments and local initiatives are playing an ordinary game where everyone knows exactly how ‘it’ is played, but without surprising outcomes.

We conclude this article by presenting two notable implications preference updating and moral disengagement, and three points about the role of citizens’ initiatives.

First, many initiators perceive the CIS-funds provided for initiatives, a serious application to receive the funding, and the accounting for the subsidy as normal and ‘part of the deal’. They do so because it is taxpayers’ money and because their initiative fits into the more general discourse of an active civil society that can count upon governmental support. These perceptions indicate some conformation to (administrative) preferences and internalised assumptions about the rules of the game that initiators and governments play. Initiators assume certain roles for the government, policy instruments, and themselves.

Second, the presence of the CIS-funds does not morally engage initiators; rather, it slightly disengages them. It disengages initiators because initiators are aware that they need money to produce the public or collective goods that they need or want. Because of this need for money, they expect the CIS-funds to be there. We also find that the CIS-funds facilitate self-determination, allowing the goods to be produced. This leads to a somewhat instrumental situation in which CIS-funds allow a logical transaction between governments asking for citizens to take responsibility, and active citizens expecting governments to enable them to do so. This lack of crowding in and the necessity of CIS-funds implies that without these funds, most initiators would not start their activities.

These two implications raise the first point of discussion about what citizens’ initiatives are an answer to. For example, can we perceive active citizens as a solutions for great societal questions (Tonkens 2006)? We saw that initiators are not actively concerned with these societal questions, but (merely and adequately) producing diverse goods because they want them: activities, gardens and playgrounds. Broader outcomes are not intentionally pursued. Prudence about what citizens’ initiatives are an answer to seems appropriate to prevent seeing citizens’ initiatives as a magic bullet that solves social, economic, and political problems all at once (Edwards 2009).

The second point is that this research has implications for how governments supposedly aim to ‘manufacture’ civil society (Brandsen, Trommel, and Verschuere 2017). Governments view citizens’ initiatives as desirable and expect them to provide innovative solutions for problems, to repair social cohesion, and to revitalise civil society. However, the aim of manufacturing or revitalising civil society through financial instruments risks at the same time destroying it. The danger is that ‘the recreated communities may not be able to exist without government support’ (ibid., 684). We have seen that initiatives are dependent on the CIS-funds, and this certainly hints at a more manufactured civil society where citizens do not self-organise in their own way, but in a way that fits the governments rules of conduct for the game of strengthening civil society.

A final point of discussion is about how CIS-funds influence civic behaviour. From our research, we can extract one optimistic and one critical implication. Starting with the former, Ryan and Deci’'s (2000a, 2000b) self-determination theory states that when people internalise regulations over time, they experience greater autonomy in action. We found initiators who responded relatively well to the fund; there were no signals of reduced self-determination. As opposed to recent findings of co-production leading to feelings of loss of autonomy (Thomsen, Bækgaard, and Jensen 2020), we found that the CIS-funds are capable of nourishing or maintaining the initiators’ motivations. On a more critical note, Bowles and Polania-Reyes (2012, 376) argue that ‘the extent to which a society relies on economic incentives – as opposed to other kinds of motivations and controls – will affect how people learn new preferences that may persist over long periods’. And while this learning is based on a long-term process and, hence, unlikely to be observed in (short-term) experiments (ibid. 375), all the initiators whom we talked to, referred to how they expect governmental action and CIS-funds and how the situation ‘just is like this’. These are all indications of conformation to preferences about policy instruments like the CIS-funds, but we could not demonstrate how these preferences have been updated or came about. On the one hand, as shown in this research, initiators may respond well to the CIS-funds and the risk of crowding out is modest. CIS-funds allow the initiators to realise their goals and produce their goods. But on the other hand, these instruments may not necessarily trigger new initiatives or more civic and cooperative attitudes. The CIS-funds have become a rather blunt instrument: they facilitate the transaction, but not the civic preferences and behaviour. Moreover, the instrumentalisation may further downplay other drivers such as compassion, patriotism, reciprocity solidarity – in other words: social capital (Bowles 1998, 92). This even further strains ambitions of revitalising a non-manufactured civil society.

Ultimately, our study has shown a more multifaceted working of crowding theory: it is possible to financially support citizens’ initiatives without crowding out high and self-determined forms of motivation. This gives hope for funding schemes to facilitate self-organisation. Highly motivated initiators can be instrumentally supported. The funds do, however, not trigger civic preferences nor behaviour.

Through our different-than-usual approach of unravelling crowding effects, we found that within the institutional context of governments supporting citizens’ initiatives there is no such fierce crowding out effects as theories generally predict. While crowding out effects were mostly absent, crowding in effects were also mostly absent. Finally, we have found clues that indicate the presence of conformation to updated preferences of initiators about their self-organisation, the availability of CIS-funds and how the game is played. Administrators and policymakers should, as such, be careful about how they deploy their policy instruments and to what ends. The fund should not only financially support, but also organise and conduct the fund in such a way that increases self-esteem and self-determination. This research has shown that CIS-funds can nourish self-esteem and self-determination to some extent: by signalling good news, by personal involvement of civil servants and aldermen, and by having ‘reasonable’ procedures.

This study has several limitations and we will name three. First, we showed nothing about the extent or the magnitude of the marginal crowding effect (e.g., does more subsidy crowd in/out more?). Second, although we have a selection of initiatives which did get financial support, we did not compare them to similar initiatives which applied for the fund and did not get the subsidy. As such, we were not able to make comparisons about types of motivation, esteem, determination, and the mechanisms between funded and non-funded initiatives. Third, we were not able to include funds that were more bureaucratic and more controlling (by means of conditions, procedures and accountability) which could be out there. Whereas these funds could trigger (fiercer) crowding out effects, our studied CIS-funds were mostly seen as relatively personal and supportive.

This article sheds light on an interesting part of the governance of citizens’ initiatives: the extent to which governments can stimulate them by means of financial incentives. Our study reveals that the relation between governments and initiatives is strongly driven by a transactional logic instead of a civic logic. In order to discover what appeals to civic behaviour, further research should uncover how governmental support can foster active citizenship and contribute to a non-manufactured civil society. Comparing whether moral engagement – of citizens and governments – could be fostered by different (or a mix) of instruments is another promising avenue for future research. Finally, future research should more thoroughly study, experimentally or qualitatively, how preferences regarding civic behaviour are affected by policy instruments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/03003930.2022.2087061.

Additional information

Notes on contributors

S. N. Blok

S. N. Blok is a PhD candidate at the Department Public Administration & Sociology at Erasmus University Rotterdam. His research interests are the collective action problems, crowding effects and value perceptions of self-organizing collectives of citizens.

H. J. M. Fenger

H. J. M. Fenger is Full professor Institutional Policy Analysis at the Department of Public Administration & Sociology at Erasmus University Rotterdam. His research focusses on analysing and explaining changes in the policies, governance, administration, and implementations of European welfare states

M. W. van Buuren

M. W. van Buuren is Full Professor of Public Administration at Erasmus University Rotterdam and Director of the Department of Public Administration & Sociology. His research interests are on issues of (co-)design for policy and governance, invitational governance and self-organization, collaborative governance and co-creation, policy innovation and institutional change.

Notes

1. Examples are ‘De Doe-democratie’, ‘In actie met de burgers’, ‘InitiatiefRIJK’, and ‘Democratie in Actie’.