Two justifications

Atkinson believes that PI can reduce the dependence on means-tested benefits. Means-tested policies, he says, produce counterproductive behavioural effects by penalising personal effort and contributing to the creation of an “unemployment trap”. Under means-tested schemes, welfare recipients may be trapped in poverty because they have limited incentives to seek and accept a job. That is because those who want to return to work may risk losing their welfare benefits once they accept a job. PI’s compatibility with other income sources undercuts unemployment’s inherent value (Atkinson Reference Atkinson1995b, 3). Thus, it should help low-income individuals improve their material situation through their own efforts. It should also improve the economic independence of married men and women who are not in paid work and who live with a partner.

As De Wispelaere and Stirton (Reference De Wispelaere and Stirton2007, 524) suggest, PI seems to be the middle ground between, on the one hand, means-test and targeted redistributive policies and, on the other hand, universal and unconditional schemes. PI inherits the contributive component of the former. As with workfare policies, PI is conditional on an individual’s behaviour. PI shares the universalism of the latter. The income entitlement for most (all adult, able-bodied) individuals is conditional on their contribution.

PI has been justified on two main grounds. The first justification is based on the assumption that PI and UBI share key normative implications but that the work requirement associated with the former will facilitate its implementation. Thus, the first argument in favour of PI over UBI is that the former is considered to be more politically feasible than the latter and normatively less objectionable than workfare programmes. As Robert Goodin (Reference Goodin2003, 74) says, socially useful participation is “politically required by mean-minded electors”. Likewise, Atkinson (Reference Atkinson1995a, 301) believes that, in order to secure political support, UBI advocates should compromise “not on the principle that there is no test of means, nor on the principle of independence, but on the unconditional payment”. If it is true that a major reason for opposition to UBI lies in its lack of conditionality, then PI should be able to secure the needed political support.

Note that there are two possible versions of the political feasibility argument. On the one hand, this argument can be grounded in the assumption that PI is a better policy, given that a UBI scheme is politically unfeasible not only in the short run but also in the long term. On the other hand, we might argue that in current circumstances PI is a more feasible policy, because it can serve as an intermediate step to the future implementation of unconditional schemes. Its conditional nature makes it easier to put together political coalitions that support implementing this type of redistributive policy. It is more likely that advocates of more extensive redistributive schemes will be better able to construct coalitions that support conditional rather than unconditional policies. However, it is an empirical question as to whether the conditional character of PI makes a significant difference in terms of political feasibility.

Similarly, we may think that the conditional nature of PI can help address and solve problems of stability. Although there are different ways to interpret the notion of stability, the most common notion utilised in the social science and normative debate is that of stability as resilience (Hansson and Helgesson Reference Hansson and Helgesson2003). When a particular variable is exposed to disturbances, we can say that that variable is stable if it returns to its original state or to a very similar state after the perturbation (Hansson and Helgesson Reference Hansson and Helgesson2003, 222). In this view, stability is related not only to the changes that take place but also to how the relevant variable would respond to potential disturbances. The stability of redistributive institutions depends highly on people’s motivation. In the context of redistributive policies, this mainly means that the redistribution of resources will not produce motivational obstacles that could eventually affect the stability of that redistributive scheme. In contrast to unconditional income guarantee programmes, we could reasonably argue that PI is likely to be more stable, because it should not affect people’s incentives to work and search for paid jobs. Unconditional schemes like UBI do not adopt any strategies to productively deal with those motivational obstacles that might affect their stability in the long term.

UBI schemes are designed to give people the freedom to decide whether or not to participate in the job market and contribute to society by performing some type of activity that may be useful to society at large. In particular, we can expect that this type of policy will not be stable if most people minimise their time spent in formal paid employment (Midtgaard Reference Midtgaard2008; Birnbaum Reference Birnbaum2012; Pérez-Muñoz Reference Pérez-Muñoz2015).Footnote ⁴ The resources that make this kind of policy possible need to be produced by some members of society. As a consequence, the main motivational challenge for UBI advocates is to find ways to motivate people to take up paid employment when they have the freedom to withdraw from paid work. PI may be better equipped than UBI to address that problem. It offers positive inducements to motivate people to either participate in the formal employment sector or to participate in activities that are socially useful but are not market generated.

Second, there is a justice-based argument for PI. The thought is that PI can better honour some important principles of justice. It can help ensure a substantial degree of equality of opportunity and satisfy the demands of reciprocity. Some scholars endorse and justify redistributive schemes that resemble PI, because they believe that these kinds of arrangements can enforce the demands of reciprocity through the welfare system (Gutmann and Thompson Reference Gutmann and Thompson1996; White Reference White2003b). For example, White (Reference White2003b) defends a conception of economic justice based on the idea of fair reciprocity. White believes that any able-bodied member of the political community who has made a reasonable contribution to the social product could claim a share of the social product (White Reference White2003b, 63). PI works well for this purpose, because it is less vulnerable than UBI to the exploitation objection (White Reference White2003a, 149). This objection says that unconditional schemes such as UBI allow some citizens to engage in morally troubling and parasitical relationships with their fellow citizens (Elster Reference Elster1986; Van Donselaar Reference Van Donselaar2008). As UBI guarantees an income without demanding a contribution in return, people are able to “free ride on the productive efforts of others” (White Reference White2004, 62).Footnote ⁵ Thus, PIs could be understood to be redistributive schemes that promote reciprocity and deter free-riding behaviours.

An alternative justification

As the feasibility and stability argument for PI has been extensively discussed elsewhere, I do not examine it further here.Footnote ⁶ For now, I focus on an alternative and underexplored pragmatic reason to favour a PI scheme – namely, the idea that this type of programme can be designed not only to redistribute income but also to allocate human resources. The allocation of services to address unmet social needs is a missing dimension in the debate about UBI and PI. Although UBI advocates argue that the unconditionality of that policy frees people up to engage in various activities – among them, the provision of services to address unmet social needs – this claim has, as of yet, remained largely unsubstantiated in the literature.

Here, I argue that PI can be conceived as a particular type of civic service programme that aims to address and tackle unmet social needs. The set of services directed to address unmet needs ranges from neighbourhood cleanups, environmental protection and gardening projects to addressing educational needs (e.g. tutoring children, literacy instruction, computer instruction, afterschool programmes for children) and volunteering in homeless shelters. As I will explain later, the activities that count in my version of PI are defined – and delimited – in an explicit listing. Therefore, we should expect Atkinson’s PI proposal to be more open-ended than the version of PI defended in this article. In other words, depending on the number and type of activities included in the list, a PI programme conceived as a civic service can be more restrictive than what a PI’s original proposal intended.Footnote ⁷

Roughly speaking, “civic service” can be defined as an “organized period of substantial engagement and contribution to the local, national, or world community, recognized and valued by society, with minimal monetary compensation to the participant” (Sherraden Reference Sherraden2001, 2). Civic service differs from occasional volunteering, as servers participate in a long-term, intensive and structured service (McBride et al. Reference McBride, Sherraden and Menon2007, 240). That definition encapsulates a wide range of activities. Civic services commonly differ in aspects such as the role that government and other organisations play in their implementation, the profile of participants recruited and the type of strategy used to motivate people to serve. For example, although some programmes are exclusively operated by public organisations, others are managed by private and nonprofit organisations (McBride et al. Reference McBride, Sherraden, Lombe and Tang2007, 5).

Likewise, civic service can be sponsored and implemented by governments and non-governmental organisations at local, state, national or international levels. Different criteria can be used to recruit servers. For instance, the three most important domestic civilian national service programmes in the US targeted participants with different profiles and characteristics. Although the Civilian Conservation Corps (CCC) recruited nearly three million unemployed needy young men and veterans, Volunteers in Service to America (VISTA) targeted college students, graduates and residents of poor communities and AmeriCorps recruited service-oriented members from different backgrounds (Bass Reference Bass2013, 1–3). Finally, participation in civic service programmes may be compulsory or voluntary. In their global evaluation of civic service in the world, McBride et al. (2003) found that, of the 210 programmes found in 57 countries, only 4% were compulsory.Footnote ⁸

Civic service programmes are designed to provide services as different as assisting and mentoring students at different educational levels, helping people to navigate and obtain access to health services and social programmes, improving recycling practices and constructing and preserving houses. Civic service is commonly aimed to address some modalities of market and government failure – that is, to provide services that neither the market nor the state typically carry out in an effective and efficient way (Moskos Reference Moskos1988). On the one hand, some of the services provided by civic service programmes can be too expensive and administratively problematic to be implemented by government employees (Moskos Reference Moskos1988, 1). On the other hand, the decisions of entrepreneurs in the marketplace may not produce socially desirable outcomes. This might happen, for example, because low demand levels for most of those services means that their provision is not commercially viable in an open market. This limited demand can be explained by different factors. To begin with, most people who need those services may be not able to afford them. This results in these services being insufficiently lucrative for market entrepreneurs. Likewise, the low demand in the market also can be affected by people’s expectations about the positive effect that those services may have in their lives.

There are at least three main justifications for civic service programmes. They tell us why we should have this kind of programme, what kind of aims they should accomplish and what kind of benefit they produce.Footnote ⁹ The first justification suggests that civic service can be designed to address unmet social needs. In particular, it is argued that this type of programme can attend to people with needs that are not covered by either the state or the market (Boyte Reference Zelleke2011). Accordingly, civic service is an alternative way of delivering at a low cost public services that otherwise would not be available to communities without depending on direct public service employment or common forms of contracting (Frumkin and Jastrzab Reference Frumkin and Jastrzab2010, 20).

A second justification comes from the idea that civic service can shape citizens’ character (James Reference James1926; Walzer Reference Walzer1983, Reference Walzer1988; Moskos Reference Moskos1988; Barber Reference Barber2003). There are two variations of this argument. On the one hand, some argue that civic service programmes promote people’s personal growth. By serving in some types of civic service programmes, people can not only develop new skills and talents but they can also expand their knowledge and capacity to understand and appreciate the diverse ways in which people live in their society. On the other hand, civic service can cultivate citizenship by instilling in people a stronger sense of civic duty and responsibility. It is thought that this type of programme improves people’s civic competence by making them more engaged and active citizens.

Finally, a third justification suggests that civic service can promote equality and social integration by bringing together people from different backgrounds (Walzer Reference Walzer1988; Barber Reference Barber2003). In particular, the argument is that this type of programme can play an important role in forging networks and social linkages. Those ties that allow people to cooperate in order to overcome different problems are, at the end of the day, an important source of social capital. Servers and the served forge connections and relationships that help build and consolidate bonds of trust between people (Frumkin and Jastrzab Reference Frumkin and Jastrzab2010, 156).

What distinguishes these justifications from one another is who benefits from the service – that is, who the intended beneficiaries are: although the first justification conceives civic service as a policy primarily designed to help needy populations, the second and the third, respectively, suggest that the purpose of participation in these programmes is to affect the participants themselves or even the society at-large. Although these justifications are independent and have different emphases, they are commonly used together to make a stronger case for community service. In fact, most programmes have a dual focus on the servers and the served (McBride et al. Reference McBride, Sherraden and Menon2007, 6). Both servers and the served can benefit from civic service. For example, not only can services have impacts on direct and indirect beneficiaries, but civic experience can also help servers to increase their skills and sense of personal worth, their understanding of community issues and problems and their sense of civic responsibility (Perry and Thomson Reference Perry and Thomson2007, 81).

Note, however, that the relative emphases placed on servers versus the served have important implications as well. Civic service programmes that primarily promote social integration and instill civic virtue may not be appropriate to address unmet social needs. A compulsory universal military service in peacetime, for example, may improve the character of those who serve and integrate people from various social positions in the society. However, it can hardly be used to meet many of the society’s unmet needs.

More importantly, the last two justifications are commonly embedded in particular conceptions of citizenship (Gorham Reference Gorham1992). The view that civic service is normatively desirable because it promotes civic virtues and inculcates citizenship is based on a contested concept of citizenship. Not only is it debatable what we should understand as “good citizenship” and “civic virtues” but also how these values can be nurtured via civic service (Gorham Reference Gorham1990, 10). For these two reasons, in this essay, I distance myself from the last two justifications for civic service. I believe that PI can be understood as being a civic service scheme, not because of its ability to instill good citizenship but because of its potential to allocate non-human and human resources that can be used to address important social needs that remain unfulfilled.

The selectivity problem

The first objection is that PI schemes can be selectively unfair (Zelleke Reference Zelleke2005, 640–641). There are two reasons behind this objection. First, if PI recipients must work in order to reciprocate the monetary benefits they receive, it is not clear why the recipients of other fiscal benefits (e.g. mortgage interest deduction, Social Security spousal benefits, etc.) are not forced to do the same. As Zelleke (Reference Zelleke2005, 641) points out, “society engages in many forms of economic distribution that benefit classes of individuals without submitting each to a work or participation test”. Second, unless the participation in socially useful activities is universally enforced, well-off members of society may choose not to participate in the provision of services to address unmet social needs. As a consequence, PI not only allows people who do not need welfare support to buy their way out of the provision of services but it also can work as an undue inducement. The latter occurs because PI monetary compensation can induce people to participate in activities that go against their principles or better judgement. The needier the person is, the greater the difficulty to opt out of the PI scheme. This places the burden of helping the needy on those who cannot afford to pass up the monetary incentives (Fabre Reference Fabre2006, 60).Footnote ¹⁰

Let us consider these two reasons in turn. The first argument is an important objection to PI programmes based on a normative principle of reciprocity. The proposals of fair reciprocity (e.g. Gutmann and Thompson Reference Gutmann and Thompson1996; White Reference White2003a, Reference White2003b) are vulnerable to this objection. In contrast to these authors, I do not argue that PI programmes further the principle of fair reciprocity more convincingly than unconditional ones. Instead, my argument is that PI programmes can be superior to unconditional ones because they can help satisfy unmet social needs.

Note, however, that the same criticism could be applied to my argument in the following form: even if there is PI in place and it efficiently addresses unmet social needs, we presumably will still have both unmet social needs and people benefiting from other public goods without rendering any relevant contribution to address those unmet needs. Therefore, my argument is not immune from the selectivity problem either. There is, however, a possible way to avoid (or at least minimise) that normative problem. The alternative is to argue that a fair PI programme would only be possible when other redistributive programmes coexisting with the PI are also designed to respond in some relevant way to unmet social needs. In other words, this possibility implies that PI programme can only be fair if there are certain background conditions in place – in this case, that other public institutions do not allocate benefits without requiring their recipients to help alleviate unmet social needs. The unfairness of the situation lies not in the fact that PI recipients are required to address unmet social needs but that the recipients of other public benefits are not meeting their obligations of justice.

That answer, however, of course, leads us to the second reason behind the selectivity problem. This second reason is based on the empirical claim that the PI scheme will be attractive only to those who have fewer options when it comes to securing an income. Those who are not in need of welfare programmes can buy their way out of serving in programmes designed to address unmet social needs. One important normative problem is that the work required to address some of those unmet social needs can be “harsh, unpleasant, cruel and difficult to endure” (Walzer Reference Walzer1983, 165). Some of these tasks might qualify as “hard work” in the sense that people do not look for and would not choose to do them if they have attractive alternatives. As Walzer (Reference Walzer1983, 165) points out, some types of work are negative goods as they typically come together with negative outcomes such as “poverty, insecurity, ill health, physical danger, dishonor and degradation”. However, because they are socially necessary, somebody must perform these tasks. Some believe that mandatory schemes are the only way to fulfil those tasks in a fair way that does not disrespect people. The thought is that all those necessary jobs that are unlikely to attract volunteers should be provided through a mandatory civic service scheme (Walzer 1983; Fabre Reference Fabre2006). Hard or dirty work should be shared among citizens. PI’s recipients may be disrespected by mandating them to perform unwanted jobs.

The ethical worry is that PI can constitute an undue inducement, which influences people to participate in activities from which they would normally prefer to abstain. Most discussions about incentives or positive inducements attempt to determine whether the response to the offer is voluntary. The voluntariness criterion tells us that a motivational strategy is illegitimate if it violates the principle of consent (Grant Reference Grant2011, 51–53). Ideally, for example, we may think that the coercive exercise of laws derives its authority from the voluntary consent of the people. Similarly, the ideal use of persuasion and incentives is one in which equal parties can act both “autonomously and cooperatively” (Grant Reference Grant2011, 53). However, these ideal conditions are rarely met in practice. We certainly do not voluntarily agree to be regulated and affected by the numerous incentive schemes, laws and persuasion campaigns that we encounter daily.

To evaluate whether or not a person accepts voluntarily to participate in PI schemes is not an easy task. PI can be an excessive offer that distorts the decisionmaking and impairs the judgement of the poor and vulnerable. However, whether or not PI programmes can lead to this outcome is debatable.Footnote ¹¹ On the one hand, PI does not target the poor. Anyone is entitled to participate in the PI programme. We know that civic service programmes can be designed to attract participants from diverse sections of the population. For example, two of the main domestic civilian national service programmes in the United States – VISTA and AmeriCorps – have successfully recruited not only residents of poor communities but also college students and graduates. AmeriCorps’ participants in particular are diverse in race and class, and they are equally distributed among high school grads, college grads and those with some college education (Bass Reference Bass2013, 223).

This, of course, does not tell us that a PI programme will have participants equally distributed from different socioeconomic backgrounds. That would be an unreasonable goal for a non-mandatory civic service scheme. However, at least it helps to defuse the challenge that PI will draw only the most vulnerable people. At the end of the day, it is an empirical question as to whether or not PI recipients will mostly consist of members of the neediest groups in society.

On the other hand, the problem of undue inducement depends on the range of activities and opportunities promoted by a particular PI programme. The broader the range of activities, the less we should be worried about the undue inducement problem. As long as PI recipients have meaningful options to provide different types of services, we should not be worried about the undue inducement effect that a PI programme may have.

The rule precision problem

The second objection is that PI programmes are likely to perform poorly in establishing rules or criteria to distinguish useful from useless activities: that is, to set up a workable criterion of eligibility to identify the intended beneficiaries (Barry Reference Barry1996; De Wispelaere and Stirton Reference De Wispelaere and Stirton2007, 526; Raventós Reference Raventos2007, 148; Maskivker Reference Maskivker2012, 131; Vanderborght Reference Vanderborght2012, 275). Two important risks are related to this task. On the one hand, it is necessary to find out how and by whom the scope and nature of participation will be determined. There is always the risk of adopting an arbitrary criterion that harms those who depend the most on this redistributive programme. For example, an eligibility criterion that establishes as socially useful only those activities that require a high level of education and skills can be highly exclusive. On the other hand, there are two relevant tradeoffs: (1) the more inflexible the criterion, the more it will resemble a workfare programme; and (2) the more flexible the criterion, the more it will resemble UBI (Van der Veen and Groot Reference Van der Veen and Groot2000; De Wispelaere and Stirton Reference De Wispelaere and Stirton2007, 526; Vanderborght Reference Vanderborght2012, 276).

The literature on rule enforcement has shown that the type of rule adopted may have an impact on not only enforceability but also the achievement of objectives (Baldwin, Cave and Lodge 2012, 230). Different rules have different levels of enforceability. Detractors of PI programmes are right to point out that Atkinson’s original proposal lacks the clarity necessary to usefully distinguish between those who perform socially useful activities and those who do not. However, we can think in terms of more precise standards for making that distinction. For instance, a more precise rule to distinguish useful from useless activities is Model 3 (discussed at the beginning of the first section). The model establishes that “A person shall be eligible to receive a PI provided he or she undertakes activities as prescribed in the following table” (De Wispelaere and Stirton Reference De Wispelaere and Stirton2007, 229). This rule would contain a detailed table specifying the different activities that can be undertaken to qualify as a PI recipient.

A more precise rule like this, however, can have additional problems. By following Colin Diver’s (Reference Diver1983) standards for regulatory precision, De Wispelaere and Stirton suggest that, although a list of activities can help us avoid excessive ambiguity, it requires that policymakers correctly anticipate the full range of socially valuable services. The main concern is that the procedures employed in populating this list may be complex and non-transparent. Transparency is one of the variables – together with accessibility and congruence – that Diver suggests are the main elements of regulatory precision. In this context, transparency is the virtue of having a “well-defined and universally accepted meaning within the relevant community” (Diver Reference Diver1983, 67). Accessibility refers to the capacity that a rule may have to apply to “concrete situations without excessive difficulty or effort”. Finally, a rule is congruent with the underlying policy objective when the message that it communicates produces the desired behaviour.

Diver argues that the precision of any administrative rule depends on these three values. Moreover, he thinks that precise rules must always face tradeoffs that occur along the three dimensions of precision. However, not all tradeoffs are equally acceptable (Diver Reference Diver1983, 71). We need a normative criterion to compare different formulations. That is precisely the problem that De Wispelaere and Stirton found with this proposal. They argue that the proposed mix of transparency, accessibility and congruence can be normatively problematic. Although this rule can distinguish accurately between PI recipients, it only achieves that goal by increasing the cost of application and reducing transparency. Poor transparency, for example, may also increase the power of street-level administrators and allow problems of creative compliance.Footnote ¹²

These are, of course, important problems that may adversely affect a PI scheme. Conditional policies have administrative and enforcement costs that unconditional ones do not have. The important question is, however, whether or not the tradeoffs associated with regulation can be minimised and legitimised. For example, there are procedures that could be used to increase the level of transparency at the time of defining the list of socially useful activities. For instance, a list of needed services can be developed through some scheme of participatory democratic governance, such as the Participatory Budget experiences implemented in various cities around the world (Fung Reference Fung2003). Participatory budget policies allow citizens to actively participate in the public budget allocation at the municipal level (Fung Reference Fung2003; Rios and Rios Insua Reference Rios and Rios Insua2007; Wampler Reference Wampler2007). Government officials and citizens participate in decision-making processes to produce public policy solutions that are favourable to the community (Wampler Reference Wampler2007, 1).

Advocates of this type of programme believe that participatory budgets offer greater legitimacy to budget allocation decisions, help diminish clientelistic practices, serve as an educational process for both citizens and public officials and induce greater transparency in public expenditure (Rios and Rios Insua 2008, 204). Although it is beyond the scope of this article to go into the details of participatory democratic governance policies, it is important to note that the basic intuition underlying these policies is that a procedure of participatory democratic governance can be used to craft an annual list of needed services.Footnote ¹³

There are various implementation issues associated with participatory democratic schemes. For instance, the level at which the participatory scheme is implemented (e.g. municipal, national) will have some impact on the performance of the policy. Likewise, it is necessary to establish criteria to evaluate the regulation and administration of those private contributive activities (e.g. care for dependents). These are just some of the issues that increase the regulatory complexity of such a policy. In short, the problem of defining a list of approved activities can be addressed through appropriate participatory mechanisms.Footnote ¹⁴

Similarly, we can learn something about the problem of identifying socially useful activities by looking at the experience of civic service programmes designed to address unmet social needs. This policy, inaugurated during Bill Clinton’s first presidential term, created the largest civic service programme in the United States – AmeriCorps (Perry and Thomson Reference Perry and Thomson2004; Bass Reference Bass2013, 154). AmeriCorps’ main goal is to engage young Americans in the services that both nonprofit organisations and local public institutions provide to satisfy unmet social needs. This programme was specifically designed to decentralise the provision of services and minimise bureaucratic burdens. It gives the fundamental responsibility of choosing and evaluating projects to state governments and community organisations (Bass Reference Bass2013, 172). AmeriCorps provides grants to programmes that are chosen by individual states and by the Corporation for National and Community Service (a federal agency). Although state commissions delineate and set their priorities and projects, nonprofit organisations are in charge of delivering the services and executing the programmes. Thus, AmeriCorps works essentially as a grant programme that provides funds on a competitive basis. Hundreds of non-profit organisations are in charge of implementing virtually the entire programme (Bass Reference Bass2013, 172). They both identify the unmet social needs of a particular community and execute services to meet them. A PI programme could adopt a similar logic for identifying socially useful activities and encouraging people to participate in different programmes.Footnote ¹⁵

Again, my intention here is not to argue that AmeriCorps can provide an adequate model to implement a PI scheme. Rather, I suggest that there is an accumulated knowledge about the implementation of civic service programmes around the world that can be used to address some of the administrative problems associated with PI schemes. The existence of these alternatives should open new possibilities for addressing the problem of rule precision. Although these alternatives cannot solve all the problems of transparency associated with the design of the list, they can at least help minimise the costs of that process.

The problem of monitoring and control

A third objection says that PI needs intrusive mechanisms to identify those who meet the criteria and to monitor those who effectively become beneficiaries of the policy. Numerous problems are associated with the use of those supervisory mechanisms. The first is related to the idea that PI interferes with individual freedom. Conditional policies interfere with people’s freedom at a higher level than do unconditional ones. Although the former requires particular behavioural responses, the latter does not. The interesting question then is whether the interference in people’s freedom associated with PI programmes is something to worry about. The answer to that question is: it depends. As PI works with a positive inducement mechanism, people voluntarily sign on to participate. This could lead us to conclude that there are no restrictions on individual freedom because nobody is coerced to participate in a PI scheme. At least in principle, incentives should not affect our freedom to choose. However, their use is not problem-free. As a form of power, incentives may be subject to abuse and misuse (Grant Reference Grant2011, x). Thus, the main reason why PI could violate people’s freedom is because it offers benefits to people who have unequal bargaining power. When a monetary incentive is too high, an offer could interfere with people’s ability to make autonomous decisions (Schmidt et al. Reference Schmidt, Voigt and Wikler2010, 3). Thus, although this is a compelling reason to be concerned about how PI programmes are implemented, it is not necessarily a concern about PI itself.

A second problem is that the verification of compliance with the policy can be too costly, complex and cumbersome to control and manage (De Wispelaere and Stirton Reference De Wispelaere and Stirton2007, 526; Vanderborght Reference Vanderborght2012, 276). There is no doubt that monitoring and enforcing PI rules can be burdensome. However, again we can refer to the experience of different civic service programmes and other policies that have overcome those difficulties. For example, conditional cash transfer programmes (CCTs) implemented in several developing countries require beneficiaries to change their behaviour with respect to the health and education of their children (Fiszbein et al. Reference Fiszbein, Schady and Ferreira2009, 205). The goal of most CCT programmes is both to provide a basic income for poor families with children and to invest in child human capital. In order to achieve the latter goal, most CCTs are conditional on school attendance and primary healthcare utilisation. Minimum school attendance for children and the fulfilment of various health checkups (e.g. following the immunisation schedule) are conditions that can be reasonably monitored. Thus, CCT programmes have a way to promote and improve children’s care by targeting parents’ behaviour. We could imagine similar ways to monitor PI recipients whose socially useful contribution lies in providing care to children and ageing or chronically ill family members. Clearly this solution cannot be extended to deal with all forms of socially useful contribution. However, variations of it can be applied to monitor and evaluate PI recipients who are in charge of providing care to family members.

The problem of undesirable economic outcomes

A fourth objection to PI focuses on the economic consequences of this type of redistributive programme. The most important criticism is that PI could displace workers who would normally take those jobs. PI would work then as a mechanism to obtain people’s labour at lower wages than the market rate (Evers Reference Evers1990, xxxiii). In an answer to this objection, it should be pointed out that a PI programme should be designed to provide services that neither the market nor the government performs. This is, of course, highly contextual, as some tasks that are performed in Community A either through the market or through government services may be performed by neither of them in Community B. Care work is an example of a service that can be carried out through different mechanisms (public care support, private market agents, volunteers, etc.) that vary from society to society. This is true for a variety of services that can be provided through a PI programme. The key point, however, is that in some contexts many important services and activities are provided in only a limited way by either the market or the government. For instance, when the potential for profit is too small, markets may not be the most adequate institutions for addressing social needs. Moreover, even when the market provides some of those services, their costs can be unaffordable for those who need it most.

Addressing educational needs (e.g. literacy instruction, computer instruction, afterschool programmes for children), planting public gardens, performing neighbourhood cleanups and providing care for the elderly and the homeless are just some suggestions of the kinds of work that markets may not provide but that PI recipients could. In any case, the key for avoiding this type of problem is to find a proper mechanism to elaborate the list of activities that can entitle someone to PI. For example, a list that is designed via a mechanism similar to the participatory budget experiences should take into consideration the demands of union representatives and other stakeholders. As Fabre (Reference Fabre2006, 68) points out, large-scale civic service programmes solve this problem by stipulating the participation of union representatives on the boards or commissions in charge of the services. As is the case with most civic service programmes, the feasibility of PI depends on its capacity to consider and accommodate the demands of various interests groups of interference.

To this day, we do not know nearly enough about PI’s potential economic impact. In contrast to other redistributive schemes, PI programmes have barely been studied.Footnote ¹⁶ The experiences of civic service programmes can help us to better understand the main administrative and economic challenges that PI could eventually face. For example, different studies have shown that civic service programmes can be cost-effective. Communities can benefit significantly from volunteers’ services (Perry and Thomson Reference Perry and Thomson2004; Frumkin and Jastrzab Reference Frumkin and Jastrzab2010, 217). There is a need for further debate about the economic impact of PI conceived as an extended civic programme, and in particular about its dynamic interactions with the labour market.

The problem of civic service under unconditional and conditional programmes

Some UBI advocates argue that unconditional schemes provide their recipients with free time to devote to community service and other socially valuable pursuits (Pateman Reference Pateman2004; Raventós Reference Raventos2007; Standing Reference Standing2011; Maskivker Reference Maskivker2012). As Standing argues, people under a UBI scheme will gain control of time. A UBI programme that delinks income from paid labour can “give people a greater capacity to live outside the market and be under less pressure to labour” (Standing Reference Standing2011, 178). This is an important feature of UBI, because it could “increase the amount of labour by allowing people to move in and out of the labour market more easily” (Standing Reference Standing2011, 178). If UBI frees people up to engage in socially useful activities that are outside of the formal paid market, then, why should we need PI after all? The thought is that UBI would allow people to voluntarily participate in a broad range of socially useful activities. Some UBI recipients may participate in services oriented towards addressing unmet social needs. When that is the case, the radically uncoordinated UBI’s approach to civic service can be preferred to the rather coordinated mechanisms of PI.

It is an empirical question, of course, whether a coordinated mechanism intended to match PI recipients to service programmes can perform better overall than the uncoordinated mechanism of UBI. At present, we do not know enough about the comparative advantages of one over the other. In any case, the worry with a UBI scheme should not be that it promotes radical idleness rather than a broad range of voluntary activities, but instead to ask whether or not UBI can motivate people to voluntarily provide services oriented to address unmet social needs. In fact, we can even assume that UBI will minimise the level of “parasitism” or “exploitation” (very few able-bodied people will live off the labour of others) and still be worried about the problem of unmet social needs. This is a more complicated challenge than most UBI advocates seem to recognise. It may be the case that the uncoordinated actions of UBI recipients will take care of unmet social needs. However, if we are really concerned about the problem of unmet social needs – as many UBI advocates seem to be – then it is unclear why a universal income-guaranteed scheme should not be directly used to promote the delivery of some important civic services.

There is at least one important reason to question the efficiency of UBI as a decentralised mechanism that allocates services to respond to unmet social needs. This reason lies in the fact that not all the unmet needs requiring society’s attention are comparable in scope and type. Broadly speaking, some of those needs might be best addressed by uncoordinated individual actions (e.g. dependent care work), whereas others may only be solved through a coordinated and cooperative effort among several people. For instance, the implementation of afterschool programmes for children or activities of environmental remediation to create more habitable living spaces are activities that require coordination and cooperation. It is more difficult to imagine in those cases that uncoordinated behaviour can lead to the desired outcomes. UBI’s decentralised nature may have more problems than PI at the time of coordinating those efforts.

UBI’s advocates can argue that UBI allows individuals the flexibility to join different organisations that are precisely designed to provide various services. I am not arguing that this is an implausible outcome. However, it is more difficult to advance the idea that UBI can be more effective in motivating people to create and join organisations that are specifically designed to address unmet social needs. If that is the case, a centralised mechanism such as PI has an operative advantage over UBI. It will not only organise and match participants with needed services, but it can also provide the adequate infrastructure to perform the required activities. UBI’s advocates need to explain in more detail why UBI may be superior to PI in that particular aspect.

My argument does not need to assume that UBI will motivate a significant number of people to free ride on the efforts of others. It will work as long as UBI fails in supporting a broad range of voluntary activities. On the one hand, the possibility that UBI has to work as a mechanism to provide services depends on its recipients’ choices about leisure and work (Watts Reference Watts2002, 132). On other hand, it also depends on the modality of services that people are willing to provide and their capacity to coordinate efforts to provide services that cannot be individually performed.

Likewise, UBI’s advocates can argue that the design of a list of activities that qualify people to get PI will inevitably produce a tradeoff between the allocative efficiency of services and the inclusiveness of the programme. The requirements of inclusion that characterise both UBI and PI should be tradeoffs in the latter policy in favour of improving the efficiency of the allocation of services to address unmet needs. The argument assumes that a more restrictive list of activities will achieve more allocative efficiency. However, that does not need to be the case.

The list of activities does not necessarily need to be restrictive in order to better allocate services to address unmet social needs. Instead, the most important role of the list should be to organise people to perform needed services that require cooperation and coordination – that is, to regulate the provision of some activities that the community collectively considers to be highly relevant. It can be the case, of course, that an insufficient number of PI recipients are willing to perform some of the most demanding services. However, even in that circumstance, PI should be in a better position to deal with the problem than UBI. Although the latter relies on people’s willingness to participate in the provision of different services, PI can be designed to offer incentives that induce people to deliver those services that are most salient at that particular moment. For instance, PI could incorporate a system of differential payments for those recipients who are willing to engage in the provision of certain services.Footnote ¹⁷ A system of differential payments would not challenge the inclusiveness of the programme, but it could help recruit more people to provide the most needed services. That does not mean that the list should be restricted to only the most pressing activities. Rather, it just means that a more inclusive list can provide differential benefits for those recipients who are willing to perform the most needed activities. Thus, PI should score better on the allocation of services compared with UBI.