by Kate McFarland | Oct 23, 2017 | Opinion
Relaxing Conditions on ‘Basic Income’: A Case Against Definition
From a linguistic standpoint, there is no one “correct” definition of the term ‘basic income’ [1]. Different groups and organizations have adopted different definitions, suitable to their purposes, and these definitions sometimes conflict with one another.
BIEN, at present, coordinates affiliates who use the term differently from one another, organizes conferences to bring together individuals who use the term differently from one another, and issues news reports on varied stories in which the term is used in different ways.
I have come to believe that, in its role as such an umbrella organization, BIEN’s attempt to define ‘basic income’ does not lend clarity. Instead, to avoid equivocation and confusion, it would do better to admit upfront this diversity in definition and shades of meaning.
BIEN’s Definition of ‘Basic Income’: One Among Many
At its 2016 Congress in Seoul, BIEN adopted the following definition: “A basic income is a periodic cash payment unconditionally delivered to all on an individual basis, without means-test or work requirement.”
When I write for BIEN’s website, I accept this as a stipulative definition of the term, and call attention to potentially confusing differences in usage. For example, when I report on the recently launched “basic income pilot” in Ontario, I note that the program being tested–in which payments to participants are household-based and income-dependent–does not actually satisfy BIEN’s definition of ‘basic income’ (although, as I emphasize below, it does congeal with an established and widespread use of the term within Canada).
My comments in might sometimes seem to treat BIEN’s definition as privileged or authoritative. This, however, is only an artifact of the particular context in which I am writing–BIEN’s website–and my desire to maintain consistency within this context. I do not believe that BIEN’s definition is privileged in any absolute or objective sense, or that it is more “correct” than other uses that have become established within other groups, organizations, and geographical regions.
As I take it, my prevailing duty as a news writer is to prevent readers from believing false things. In this context, clarity and consistency in meaning are of utmost importance, and BIEN’s definition of ‘basic income’ is a burden I must bear, knowing that there will be many situations in which it will be inconsistent with the definitions employed by the parties on whom I report.
Two particularly important cases, in my experience, are the following two types of definitions:
- Definitions that additionally stipulate that the amount of the periodic cash payment must be sufficient to meet basic living expenses.
- Definitions that lack the qualification that the payment must be (a) non-withdrawable (not means-tested) and/or (b) paid on an individual basis.
A. Definitions stipulating that the amount of the periodic cash payment must be sufficient to meet basic living expenses.
Many high-profile groups and organizations have adopted definitions of ‘basic income’ with this additional necessary condition (philosophers may enjoy the opportunity to say that, on these definitions, “the ‘sufficient’ condition is a necessary condition”); for example (emphases added):
- GiveDirectly, the charity organization known in the basic income community for its impending major experiment in Kenya, defines ‘basic income’ as a type of cash transfer that is “unconditional (recipients don’t have to work or do anything else to be eligible), universal, with all members of society receiving, enough to cover basic needs, and guaranteed for the recipients’ lifetimes”.
- International Basic Income Week, an annual initiative that is pursuing partnership with BIEN in 2018, stresses four conditions that must be met to use ‘basic income’ to refer to a cash transfer policy: payments must be (1) universal, (2) individual, (3) unconditional, and (4) high enough.
- Founding members of the Economic Security Project, a major US-based initiative launched in late 2016, have decided to reserve the use of ‘basic income’ for programs in which payments are high enough to meet basic living expenses, and have endorsed the neologism ‘base income’ to refer to programs that provide universal and unconditional payments of lesser amount.
- Multiple affiliates of BIEN–including groups in Australia, Austria, Canada, Germany, the Netherlands, Norway, Portugal, and Switzerland, and perhaps others–have adopted definitions of ‘basic income’ (or its translational equivalent) that include some type of condition specifying that the amount of payment must be “sufficient” or “high enough” to meet some type of minimal needs (see “Affiliate Definitions of ‘Basic Income’”).
The question of whether BIEN itself should include the sufficiency condition as a necessary condition (so to speak) has been the cause of previous terminological controversies within the organization, including the one that eventuated in the vote at the 2016 Congress in Seoul, in which BIEN rejected the proposal to restrict the definition of ‘basic income’ in such a manner. In a paper delivered at the 2017 BIEN Congress (“What’s a Definition? And how should we define ‘Basic Income’?”), Malcolm Torry, General Manager of BIEN and Director of the UK’s Citizen’s Basic Income Trust, defends this decision. According to Torry, the adopted definition does not “conflict with any affiliate’s definition”, “represent[s] the consensus among affiliates”, and “reflect[s] common usage of the term”.
All of these claims seem dubious, however, especially when one considers that BIEN has aspired to provide a definition–that is, a set of necessary and sufficient conditions to use the term ‘basic income’–rather than a non-exclusive list of necessary or paradigmatic features.
First, notice that the different definitions of BIEN and some of its affiliates lead to different assignments of truth and falsity to certain sentences. For example, the sentences ‘Alaska has a basic income’ and ‘Iran once implemented a basic income’ seem to be true on BIEN’s definition [2], but they are false on definitions of ‘basic income’ that include a provision that the amount must be high enough to meet basic living expenses. The definitions disagree on whether certain core cases discussed in the basic income literature are actually cases of basic income–and this, I wager, counts as “conflict” between the definitions if anything does.
Secondly, it would be more accurate to say that BIEN’s definition reflects a–but not the–common usage of ‘basic income’, and that it does not represent consensus, even amongst BIEN’s own affiliates. As reflected by the above list of examples (and further examples could be given), it is unquestionably typical for many speakers and organizations to restrict application of the term to policies that provide livable cash payments.
If one still wonders why a less restrictive definition should prove contentious, it is significant to notice that the act of defining carries evaluative and expressive element: to establish a definition of a term is not merely to clarify and elucidate current usage, nor is it necessarily an attempt to honor as much of present usage as possible while introducing greater clarity and precision; to establish (and insist upon) a specific definition is often also to express what one values. Specifically, speakers sometimes choose to adopt definitions that are narrow or exclusive and reject ones that are more encompassing.
Many conservative Christians, for example, continue to resist the redefinition of ‘marriage’ to allow the term’s application to same-sex partnerships. And many pro-choice Americans were recently outraged that the US Department of Health and Human Services decided to define ‘life’ broadly to the point of conception. Or, in a somewhat lighter vein, consider cocktail purists who scoff at the practice of using ‘martini’ to refer to any mixed drink served in a v-shaped glass. In the eyes of the conservative Christian and the cocktail purist, the more inclusive definitions are simply unacceptable, for they disrespect the sanctity of marriage and martinis (as God and the International Bartenders Association intended them to be). And when the definition in question has legal or political ramifications, the choice bears substantial weight.
Likewise, in my experience in the basic income movement (especially in the US context), I have observed that many left-leaning proponents of basic income insist upon the strict definition–with the “sufficiency” condition–as a way to distance their own proposals from right-wing and libertarian schemes, such as Charles Murray’s proposal to replace all existing programs with a universal flat-rate cash payment of 10,000 USD per year. Often, champions of a narrow definition of ‘basic income’ don’t want Murray-like proposals to be assimilated into the basic income movement, and their preferred definition reflects this.
Such activists might decry definitions like BIEN’s as unacceptable precisely because it aspires to retain neutrality on the level of the payment. If BIEN were simply to reply that the definition should be kept broad in order to accommodate all proposals for regular unconditional cash transfers, including views like Murray’s, then it would quite directly miss their point. It is their prerogative as speakers to fine-tune the meaning of terms in light of their values and interests, and they might have good strategic and political reasons to insist upon these particular definitions.
I believe that BIEN should acknowledge this difference as what it is: disagreement about word meaning–and not at all uncontentious. This disagreement threatens to present readers and newcomers with potentially confusing discrepancies, such as disagreement about the truth of such commonly heard claims like ‘Alaska has a basic income’ and ‘Finland is experimenting with a basic income for its unemployed population’. For the reader of BIEN materials, forewarned is forearmed.
B. Definitions lacking the qualification that the payment must be (a) non-withdrawable (not means-tested) and/or (b) paid on an individual basis.
As mentioned above, this type of definition seems particularly common in Canada, where the idea has enjoyed a long history, where an experiment in Manitoba in the late 1970s became one of the most famous trials of a negative income tax–or what many Canadians politicians, academics, and journalists would call a ‘basic income’.
For example, on a page titled “About Basic Income”, BIEN’s Canadian affiliate, Basic Income Canada Network, defines ‘basic income guarantee’ (which seems to be used synonymously with ‘basic income’) as a program that “ensures everyone an income sufficient to meet basic needs and live with dignity, regardless of work status” [3] [4]. Similarly, the Government of Ontario, which has recently launched a new experiment of what it calls a ‘basic income’, defines the term as “a payment to eligible families or individuals that ensures a minimum income level, regardless of employment status” [5].
Exemplifying this usage, Canadian politician Guy Caron has introduced a proposal for what he calls ‘basic income’, which is a “top-up aimed at helping low-income Canadians to reach the ‘low-income cut-off’”, clearly not a universal and non-withdrawable payment. I submit that this proposal, like Ontario’s pilot, is not inaccurately named: it merely reflects what might be described as dialectical ambiguity with respect to the term ‘basic income’.
Also in his 2017 paper, Torry states, speaking of the Ontario experiment, that the payments “do not constitute a Basic Income, and perhaps BIEN should say that”. I would contend the appropriateness of Torry’s advice depends, in part, on one’s audience. If addressing an audience of basic income activists in UK, for example, then it might indeed be important to clarify that Ontario’s pilot is “not a basic income” (assuming they endorse, and are most familiar with, the definition of ‘basic income’ adopted by BIEN, the Citizen’s Basic Income Trust, and Basic Income UK). But it would be quite presumptuous to make the same assertion to the Government of Ontario itself.
By analogy, suppose a British child were to overhear an American speak of “eating biscuits with dinner before the football game”. It might be important to clarify that the food in question is “not really biscuits” and the game in question is “not really football” in order to prevent the child from forming misconceptions about the American’s selections of baked goods and sports. But it would not be appropriate, presumably, to tell the American himself that he is “not really eating biscuits or watching football” and should stop using his words like that. But Ontarians are not en masse misusing the term ‘basic income’ any more that Americans are en masse misusing the words ‘biscuit’ and ‘football’. The term has merely taken on a different meaning, one with antecedents dating at least to the time of the Mincome experiment in the 1970s.
Given this divergence in meaning, there is again a pronounced threat of equivocation and confusion. Widely used sentences like ‘Ontario is testing basic income’, ‘Manitoba’s Mincome was an experiment of basic income’, and ‘Milton Friedman supported basic income’ might be either true or false depending on the speaker. Indeed, the truth or falsity of such sentences can be determined only after knowing the specific definition of ‘basic income’ adopted by the speaker (or, as a clue, the speaker’s nationality).
Once again, I believe the lesson here is that a wide-scope organization like BIEN must acknowledge this diversity in word use if it wishes to ward against such confusion.
C. “Similarities Overlapping and Criss-Crossing”
In working as a reporter for Basic Income News, I commonly observe speakers–many of them with considerable experience and expertise in the movement–use the term ‘basic income’ in accordance with the two types of definitions described above.
In itself, such ambiguity is benign; it is a common feature of natural language that the meanings of words are shaped and honed in somewhat different fashions within different communities or groups of speakers. When a speaker realizes that a term carries multiple definitions, she knows that she must attend to context in order to resolve the ambiguity, and she knows to be cautious of drawing certain inferences if that ambiguity cannot be resolved.
Complications arise, however, when casual readers falsely assume that ‘basic income’ is well-defined and unambiguous. And, unfortunately, this is all too easy: most articles and websites that purport to introduce “the” concept of basic income do not mention that the term is used differently, and sometimes inconsistently, between different speakers. On the contrary, many authors blithely write as if the term does have a single conventional meaning, offering a gloss on a definition with no mention of the fact that others define the term somewhat differently. Thus, many casual readers might be unaware of the ambiguities that surround the use of the term–raising the specter of unintentional equivocation, confusion, and false belief (e.g. one might unwittingly come to accept falsehoods like “Alaska provides its residents with livable annual income” or “A town in Manitoba was the site of an experiment in which every resident, regardless of income, received a fixed monthly cash payment”).
If an organization like BIEN wishes to be a broad church, facilitating discussion between diverse parties that research or support something they call ‘basic income’, then, I submit, it should cease to posture as if the term has a single definition.
BIEN could state outright that there is no set of conditions that constitutes a unique standard meaning for ‘basic income’. Then, as an alternative to definition, it could provide a list of stereotypical or paradigmatic features of proposals that bear the name. Some of these features–such as being paid in cash and at regular intervals–are more central than others, and might even be said to be necessary features of anything called ‘basic income’. Other features, however, are matters of dispute or discrepancy (such as consisting of a livable amount and being paid in equal amount to all regardless of income).
New problems would likely arise when attempting to decide what to say about frequently cited conditions or features of a basic income. Even the condition of unconditionality, for instance, might not be sacrosanct. Some have spoken about questions such as whether a “participation requirement should be added to the basic income”: is this loose talk, semantic nonsense (akin to asking, perhaps, whether a “marriage requirement should be imposed on bachelorhood”), or evidence that unconditionality is not really a necessary or immutable part of the conceptual core of what speakers call ‘basic income’? If informed and competent speakers’ judgements fail to detect paradox in phrases like ‘a participation requirement on a basic income’, then it is likely the latter.
Yet more new problems would arise when considering the fact that the paradigmatic form of the policy contains further attributes that are often not mentioned explicitly in definitions of the term: the amount is typically assumed to be relatively stable; the program is typically assumed to be created and administered by a government (although, with some activists proposing privately-funded programs, some definitions deliberately reject it); the condition of “universality” is often (although not always) implicitly assumed not necessarily to extend to children. (See Torry’s paper for further discussion of examples of implicitly accepted paradigmatic features.)
Still more points of controversy could be mentioned. Should ‘basic income’ be defined to require payment in conventional currency (to exclude cryptocurrency-based proposals), or should it be sure not to impose this constraint? Should ‘basic income’ be defined as a universal payment to citizens, permanent residents, or some other specification of the relevant population base?
At the beginning of his 2017 Congress paper, Torry mentions Ludwig Wittgenstein’s discussion of the idea of “family resemblance” in his Philosophical Investigations. He does not, however, carry this Wittgenstein reference to its natural conclusion: a case against definition. I suggest that we do.
According to Wittgenstein, terms of natural language typically do not lend themselves to definition in terms of necessary and sufficient conditions. Different uses of a term need not exemplify a common core meaning–and, often, they don’t. Instead, Wittgenstein tells us, different uses of a word are often related by a “complicated network of similarities overlapping and criss-crossing”.
To illustrate, he delivers the example of the word ‘game’:
Consider for example the proceedings that we call “games”. I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all?—Don’t say: “There must be something common, or they would not be called ‘games’ “—but look and see whether there is anything common to all. … Look for example at board-games, with their multifarious relationships. Now pass to card-games; here you find many correspondences with the first group, but many common features drop out, and others appear. When we pass next to ballgames, much that is common is retained, but much is lost.—Are they all ‘amusing’? Compare chess with noughts and crosses. Or is there always winning and losing, or competition between players? Think of patience. In ball games there is winning and losing; but when a child throws his ball at the wall and catches it again, this feature has disappeared. Look at the parts played by skill and luck; and at the difference between skill in chess and skill in tennis. Think now of games like ring-a-ring-a-roses; here is the element of amusement, but how many other characteristic features have disappeared! And we can go through the many, many other groups of games in the same way; can see how (§66).
The natural progression of Torry’s Wittgenstein reference would have been to argue that ‘basic income’ is like ‘game’: as we examine the diverse and multifarious uses of ‘basic income’, similarities crop up and disappear, with no single common meaning able to be identified. Although Torry does not take this tack, I believe that it would have been precisely on-point.
Of course, an association of “gamers” is free to stipulate a specific definition of the particular type of game in which it is interested. Likewise, an organization like BIEN could stipulate a specific definition of ‘basic income’ to describe the particular type of policy with which it is concerned. One concern, however, is that BIEN’s other actions don’t seem to accord with this desire for specificity. BIEN, at present, seems unified more by word-shape than word-meaning: it coordinates affiliates who support what they call ‘basic income’ (allowing affiliates to adopt their own definitions thereof), organizes conferences to bring together individuals who are interested in something they call ‘basic income’, and publishes news stories about people who talk about something they call ‘basic income’. If BIEN wishes to unify its activities in this way, then it cannot prescribe a definition of ‘basic income’ but must instead defer to the groups and individuals who use the term and who constitute its membership–and, as seen, their definitions of the term are varied and disparate, with only a thin and insubstantial core of features possessed by all.
But we may leave aside questions of BIEN’s mission and goals, for there is another concern facing the organization’s decision to adopt a particular stipulative definition: as a mere matter of fact, the term is used in varied manners that are not always consistent, media coverage of basic income is not always clear to dissociate these (and is usually not), and BIEN does not hold purchase over media reporting on basic income. Adding yet another organization-specific definition to the mix does not lend clarity to confusion. What is needed is straightforward acknowledgement of the diversity and disparity in uses of the term ‘basic income’.
Notes
[1] In this essay, I use single quotes to notate that I am speaking about a linguistic item (the term ‘basic income’) rather than the thing it refers to (a basic income).
[2] Alaska’s Permanent Fund Dividend and Iran’s oil subsidy reform program possess other non-stereotypical features. Most notably, perhaps, the level of the payments is not only non-livable but also unstable and uncertain. No fixed amount is guaranteed from year to year. (Indeed, as it happens, the future of the PFD is presently uncertain due to ongoing fiscal crisis in Alaska–in an unprecedented decision of the state’s Governor, its amount was halved between 2015 and 2016–and Iran has begun withdrawing higher earners from the subsidy program.)
[3] Note that other groups, such as BIEN’s Australian affiliate, also use ‘basic income guarantee’ and ‘basic income’ interchangeably, but with a definition that includes the qualifications that the payments must be individual and non-withdrawable; thus, the use of the word ‘guarantee’ does not imply that the “Canadian-type” definition of ‘basic income’ is at play.
[4] BIEN’s US affiliate, the US Basic Income Guarantee Network, defines ‘basic income guarantee’ in a manner similar to BIEN’s Canadian affiliate, as “a government ensured guarantee that no one’s income will fall below the level necessary to meet their most basic needs for any reason” (with no condition that the support must be non-withdrawable or paid on an individual basis). However, USBIG does not treat ‘basic income guarantee’ and ‘basic income’ as synonyms, but defines ‘basic income’ in a manner similar to BIEN, as type of basic income guarantee in which “every citizen [is given] a check for the full basic income every month” (leading to peculiarities like the truth of the sentence ‘Ontario is testing a basic income guarantee, but it is not testing a basic income’).
[5] Although common, this is not the only definition in use in Canada. For example, François Blais, a political scientist researching income guarantee programs for Quebec, has defined ‘basic income’ as “an unconditional income that the government awards to every citizen” (see his book Ending Poverty: A Basic Income for All Canadians)–which, as written, could be interpreted as implying an individual and non-withdrawable payment (although not explicitly specified).
Earlier draft reviewed by Tyler Prochazka and Heidi Karow
Cover Photo (Games): CC BY-NC-ND 2.0 B
by Kate McFarland | Oct 19, 2017 | News
Existing and Upcoming BI-Related Experiments
By Kate McFarland
Last updated: October 15, 2017
It seems that 2017 has been a watershed year for the global basic income movement, as multiple governments and private research groups have independently conceived and launched experimental trials of basic income (and closely related policies). Several new experiments in North America and Europe represent the first such experiments in the developed world since the 1970s (when a negative income tax was tested in several cities in the United States and Canada), and the largest basic income trial ever designed is about to take place in Kenya.
At the same time, rumors of other experiments have appeared only to be revealed to have been premature, and sloppy and superficial media reports have obfuscated differences in the design and motivation of these disparate studies. This article reviews the latest information (as of October 2017) on the experiments that actually are being conducted (or well along the path) [1] [2].
The federal government of Finland is currently conducting an experiment of the effects of a basic income on unemployed citizens, which began in January 2017 and will conclude in December 2018. Prior to the launch of the Finnish experiment, the provincial government of Ontario had already announced its plans to test a type of unconditional income guarantee; at the time of this writing, it is currently enrolling participants in three areas of the province, who will receive an income guarantee for up to three years. In the Netherlands, another instigator of the recent interest in basic income experiments, municipal-level experiments have faced setbacks and changes in the quest to meet compliance with federal law; however, as of October, several cities have now launched experiments with the removal of conditions on social assistance benefits. The city of Barcelona has launched an experiment testing several potential reforms of its anti-poverty programs, including new social programs as well as unconditional cash payments. Proposals to test basic income at the municipal level have also lately gained considerable political support in Scotland.
In addition, two US-based non-profit organizations have completed pilot studies, and are preparing to launch privately funded basic income experiments on a large scale. After the Kenyan elections, the charity GiveDirectly plans to initiate a 12-year randomized controlled trial (RCT) testing the effects of universal basic income on villages in rural Kenya. Meanwhile, Silicon Valley’s Y Combinator Research has completed a feasibility study in Oakland, California, and is now finalizing the design of an RCT that is to involved 3,000 participants in two US states. (Edit: A privately funded trial in Stockton, California–announced around the same time this article was published–may soon join the list. However, the project is in its early planning stages, and few details have yet been determined.)
1. Finland’s “Perustulokokeilu” (Basic Income Experiment)
Status: Launched on January 1, 2017; in progress until December 2018.
Official website: https://www.kela.fi/web/en/basic-income-experiment-2017-2018
In October 2015, the federal government of Finland formed a working group to research the design and implementation of a nationwide basic income experiment, described as a means to “find ways to reshape the social security system in response to changes in the labor market”.
One prevailing concern with existing systems of social security was the steep rate at which benefits are clawed back when beneficiaries receive a job, which has been hypothesized to discourage job-seeking. Additionally, those who gain employment on a short-term basis need to reapply for benefits after their position ends, often resulting in a gap in financial support. The latter has become a particular concern due to the increase in precarious work, such as temporary and contract positions (see, e.g., Marjukka Turunen’s presentation to Kela). Thus, the idea of unconditional basic income gained attention as a possible means to remove practical and psychological barriers that might currently deter unemployed Finns from looking for work.
After reviewing several design suggestions proposed by Kela, the Social Insurance Institution of Finland, the government settled on the program that was launched on January 1, 2017. The experimental group consists of 2,000 persons, who were randomly selected from Finns between the ages of 25 and 58 who had been receiving unemployment benefits from Kela in November 2016. (The remainder of the sample population, totaling around 175,000 individuals nationwide, constitutes the control group.)
These 2,000 participants are receiving unconditional monthly cash payments of €560 (about 590 USD), an amount insufficient to meet basic living expenses, but approximately equal to that provided by Finland’s existing programs of unemployment assistance.
In contrast to those who continue to receive Kela’s existing unemployment benefits, participants in the basic income pilot are not required to demonstrate that they are seeking employment, nor are they required to accept jobs offered to them, and those who do obtain work will continue to receive the full benefit.
The first payouts to experimental participants were distributed on January 9, 2016, and payments will continue through the end of the trial in December 2018. To avoid selection bias, participation was mandated for those chosen, and exit from the experiment is prohibited. However, the experiment has been designed to ensure that no participants will be placed in a worse financial position than they would have experienced under their previous benefits.
Labor supply effects are the main outcome of interest: the experiment will assess whether the final unemployment rates differ significantly between those individuals receiving the basic income and those receiving traditional employment benefits. Kela has also stated plans to examine difference in expenditure on medication, health care usage, and income variation.
To avoid observer effects, Kela aims to minimize interaction with experimental subjects during the duration of the trial, and will conduct no surveys or interviews of subjects until the experiment has concluded. Most analysis will be based on registry data that can be obtained without direct interaction with participants. No results of the experiment will be released prior to 2019, after the period of data collection has concluded. (Rumors of early results demonstrating a decrease in stress level were, in fact, based on a single anecdote voluntarily divulged to media.)
While many basic income supporters and “BI-curious” individuals have praised Finland’s initiative launching the first nationwide experiment of a basic income program, many activists have also expressed disappointment with the final design of the experiment, questioning its ability to produce useful results and even whether it should be called a “basic income” at all [3]. Particularly controversial have been the decisions to test only a “partial” basic income (i.e. an amount insufficient to meet basic living expenses), limit the target population to those who had previously been receiving unemployment benefits, and focus primarily on labor supply effects.
Research director Olli Kangas has recommended expansion of the experiment in future years in order to test different models of a basic income or broaden the target population. This, however, will depend on budgetary decisions of the federal government.
2. Ontario’s Basic Income (Guaranteed Minimum Income) Pilot
Status: Officially announced in April 2017; currently enrolling participants.
Official website: www.ontario.ca/page/ontario-basic-income-pilot
Before describing Ontario’s “Basic Income Pilot” (as the study is officially called) on BIEN’s website, a word about terminology is in order: in the Canadian context, the term ‘basic income’ is commonly used in a more expansive manner than the definition adopted by BIEN to refer to programs that guarantee minimum income, with no type of work requirement, for all members of society without any type of work conditions. This is more expansive that BIEN’s definition in that it omits the conditions that policy must provide payments to individuals (rather than households) and in an amount not dependent on additional earned income.
For example, Canadian writers and policymakers often use ‘basic income’ to refer to programs in which cash payments are not distributed universally, such as a negative income tax or top-up of low incomes. (In common Canadian terminology, the word ‘demogrant’ is used equivalently to BIEN’s use of ‘basic income’, and to refer to a sub-type of the programs that most Canadians call ‘basic income’.)
River in Lindsay, Ontario, CC BY 2.0 RichardBH
This clarification is important given that, in fact, Ontario’s experiment employs cash transfers that depend in their amount on both income and household status. Specifically, single participants will receive a guaranteed annual income of 16,989 CAD (€11,340), while couples will receive a minimum of 24,027 CAD (€16,038) per year (amounts pegged to 75% of the Low Income Measure or LIM, where the LIM is roughly 50% of area median income).
Moreover, the above amounts are the maximum payments; that is, they are the amounts that would be paid out to individuals and couples with no external income source. The amount of the benefit will be reduced by the amount of 50% of any earned income (e.g. if a single individual in the study receives 20,000 CAD per year in earned income, she will receive an additional 6,989 CAD per year through the pilot program, or 16,989 CAD less 10,000 CAD). This entails, for example, that single individuals will not receive any payment through program if their annual earned income rises above 48,054 CAD.
Additionally, only Ontarians with an annual income below a certain level (34,000 CAD for single individuals or 48,000 CAD for couples) are eligible to participate in the experiment. (In contrast to the experiments in Finland, the Netherlands, and Barcelona, the target population is not restricted to current welfare recipients; it is, however, restricted to the low-income population.)
Although the amount of the cash supplement depends on income and household status, it does not depend on employment status, participation in job-seeking activities, training, or any other prescribed activity, or proof of an inability to work (although individuals with disabilities can receive an additional amount of up to 500 CAD per month).
Compared to the Finnish government, the government of Ontario is less focused on employment effects in particular, and more interested in the ability of a guaranteed income program to reduce poverty, food insecurity, and mental and physical health problems caused or exacerbated by low or unstable income. This is one reason that, in the context of the Ontario experiment, it may be less significant that the cash benefit is clawed back with earned income; researchers and policymakers are less focused, if at all, on reducing or removing the “benefits cliff”. Meanwhile, poverty–the prevailing concern to Ontario–is not an issue motivating the trial in Finland.
Thunder Bay lighthouse, CC BY-NC 2.0 C Hanchey
Since June 2017, the Ontario government has been enrolling participants from the three regions chosen as sites for the experiment: the Hamilton, Brantford, and Brant County region, Thunder Bay and surrounding area, and the city of Lindsay.
Residents of these regions have been randomly selected to receive application packages, and are eligible to enroll if they are between the ages of 18 and 64 with income below the levels mentioned above. Unlike the situation in Finland, participation is voluntary, and participants may opt out of the experiment at any time.
At the time of this writing, 400 participants have been enrolled from Hamilton and Thunder Bay areas. The government intends to enroll at total of 2,000 participants from these two regions, in addition to 2,000 from Lindsay (where enrollment will begin later in the year).
Participants will be regularly surveyed about topics such as their health, employment, and housing situation. Those assigned to the control group will receive no cash benefit but will be administered the same surveys. A third-party research group will evaluate outcomes in a variety of areas, including food security, stress and anxiety, mental health, health and healthcare usage, housing stability, education and training, and employment and labor market participation.
The experiment will continue for three years after its launch, with payments distributed on a monthly basis, and results are expected to be reported to the public in 2020.
3. Dutch Social Assistance Experiments
Status: Two-year experiments have been launched in four cities in October 2017, after meeting compliance with federal legislation; a fifth will follow in December (Nijmegen), and two cities (Amsterdam and Utrecht) are discussing revisions necessary for legal compliance.
In contrast to the cases in Finland and Ontario, in which the national and provincial governments (respectively) have called for and overseen the implementation of the trial programs, social assistance experiments in the Netherlands have developed “from the bottom up”: municipal leaders and university researchers in several Dutch cities planned experiments to test the replacement of the nation’s “workfare” benefits with unconditional cash assistance, and sought permission to carry out these experiments under the auspices of the Participation Act.
This national law, passed in 2015, tightened conditions on the receipt of welfare benefits, with the goal of promoting reintegration into the labor market. For example, individuals are commonly required to complete five job applications per week, attend group meetings, and participate in training activities in order to continue to receive their benefits. However, the law also granted municipalities the opportunity to implement new forms of social assistance on a trial basis for up to two years–subject to certain constraints (to be discussed below). Hence, the Participation Act might be said to have provided both the impetus to experiment with basic income (as an alternative to its workfare-based to benefits) as well as the license to do. As will be seen, however, the conditions of this legislation have also effectively precluded the ability of researchers to test a truly unconditional and non-means-tested basic income.
The Dutch municipal experiments, like the Finnish experiment, were motivated in large part by concern about the ability of existing welfare programs to incentivize work. However, while the Finnish experiment focuses on the role of monetary incentives (e.g. the ability to retain cash benefits after taking a job), the Dutch experiments have also stressed individuals’ intrinsic motivation to work. For example, drawing upon work in behavioral economics, Utrecht University economists Loek Groot and Timo Verlaat have argued that coercing individuals to work, as in the case of the nation’s “workfare” programs, can undermine their intrinsic motivation to perform fulfilling work and make productive contributions to society.
Thus, although the Dutch municipal experiments are designed to investigate the effect of removing financial disincentives to work (reducing the withdrawal rate on means-tested benefits), as described below, the most discussed experimental intervention has been the removal of coercive reintegration requirements on welfare recipients.
Researchers at the Universities of Groningen, Tilburg, and Utrecht initially proposed to test a policy close to a basic income. However, research teams experienced setbacks in 2016 and 2017, as interaction with the national Ministry of Social Affairs and Employment made clear that the terms of compliance with the Participation Act would prohibit experimentation with a truly unconditional benefit.
For example, if an experiment involves the removal of work reintegration requirements on benefits, the Ministry mandates that municipal officials survey test subjects after six and twelve months to verify that they have made adequate effort to find employment: if any participants are determined not to have made adequate job-seeking efforts during this time, they are subject to dismissal from the experiment. The Ministry has also required that any experiment including a treatment group with relaxed conditions on the receipt of benefits also must include a treatment group with stricter conditions (such as more intense reintegration activities). Finally, the Participation Act caps the amount of earned income that participants are permitted to retain on top of their social assistance benefits at €199 per month, meaning that a completely non-withdrawable (non-means-tested) benefit cannot be tested.
Despite such setbacks and constraints, however, several municipalities have designed–and are beginning to launch–experiments that meet the conditions of the Participation Act.
On July 3, 2017, the Dutch Ministry of Social Affairs and Employment granted permission to four municipal-level social assistance experiments–in Groningen (with the neighboring village of Ten Boer), Wageningen, Tilburg, and Deventer. An application from Nijmegen was also approved shortly thereafter. Meanwhile, discussions are ongoing in Amsterdam, where the city has so far refused to adopt the stricter conditions required by the Participation Act, and Utrecht, where an application to experiment was approved conditionally on revising a city law to meet compliance with the act (to be decided in December).
In each of the two-year experiments, participants will be (or have been) randomly selected from a pool of current social assistance beneficiaries, and assigned either to a control group or to one of several treatment groups.
As in Ontario’s experiment, but in contrast to Finland’s, participation is voluntary for those selected.
The experiments in Tilburg and Wageningen include three treatment groups: (1) a group with the removal of reintegration requirements, such as job application quotas and participation in training programs, on welfare benefits; (2) a group with a more intensive form of reintegration service for welfare recipients (as per the requirements of the Participation Act); (3) a group permitted to keep additional income earned on top of welfare benefits (subjects in this group may retain 50% of additional earned income, up to the mandated maximum of €199 per month, for the duration of the two-year experiment; under current policy, welfare beneficiaries are permitted to keep only 25% of additional income, and only for up to six months).
The Groningen / Ten Boer experiment includes a fourth treatment group, in which participants are permitted to choose to join any one of the three preceding groups.
The Nijmegen experiment, in contrast, will combine features of above treatment groups into two group. In one group, reintegration requirements will be removed; in the other, more intensive reintegration requirements will be implemented, but subjects will be allowed more autonomy and discretion in selecting reintegration activities (e.g. volunteer work, help obtain full-time work, or assistance for entrepreneurship). In both, subjects will be able to retain of 50% of additional income (up to the €199 maximum).
Researchers plan to examine outcomes such as health, stress level, subjective well-being, financial well-being (such as amount of debt), education, employment (including part-time and temporary employment), and participation in social and cultural life.
The experiments in Tilburg, Wageningen, and Deventer began at the start of October 2017, with Groningen / Ten Boer to follow at the end of the month. Nijmegen plans to launch its experiment in December.
4. Barcelona’s B-MINCOME
Status: Launched in October 2017
Official website: https://ajuntament.barcelona.cat/bmincome/
Launched in October 2017, Barcelona’s B-MINCOME experiment is exploring several potential solutions to address poverty and social exclusion. The experiment is being conducted in Besòs area, the city’s poorest region, and, as in the above experiments, the target population consists of low-income individuals and households. As in the Finnish experiment and Dutch municipal experiments, participants are drawn from current recipients of social assistance benefits (in this case, Barcelona’s Municipal Social Services). Once again, although the program is thereby not a test of a truly “universal” benefit, this restriction is sensible in the context of the experiment, the main objective of which is to test the effectiveness of alternative anti-poverty programs.
A stratified random sample of 2,000 households was selected for the experiment, with 1,000 households assigned to the control group, and the other 1,000 assigned (at random) to one of ten treatment groups. (As in the Ontarian and Dutch experiments, but in contrast to the Finnish experiment, participation in the experiment was made voluntary for the selected households.)
All of the treatment groups will receive cash income supplements (called “Municipal Inclusion Support” or, in the Spanish acronym, “SMI”), but differ according to whether the SMI is accompanied by an additional social program and whether the SMI is means-tested. In general, the amount of the SMI will depend upon household composition and financial status, and is expected to range from €100 to €1,676 per month per household. A total of 550 households in the experiment will be assigned to participate in one of these four social programs, including an occupation and education program, a social and cooperative economy program, a guaranteed housing program, and a community participation program. The remaining 450 households will receive the SMI without any associated programs. Within this group, the receipt of the SMI is not conditional on work, willingness to work, or willingness to participate in any other type of program. Furthermore, for some of these 450 households, the SMI will not be means-tested.
Hence, although B-MINCOME is not only a test of basic income–it is also a test of the effectiveness of the additional social programs–it includes a trial of basic income. This is not coincidental: the project team contains several members of BIEN’s Spanish affiliate, Red Renta Básica, and consulted with representatives from Finland, Ontario, and Utrecht who have been involved with the design of the experiments in their respective regions (not to mention that its name refers to Manitoba’s well-known experiment with unconditional income guarantee, in the form of a negative income tax, in the 1970s).
To examine the impact of the cash transfers (SMI) and social program on poverty and social exclusion, researchers will examine a number of outcome variables, including labor market participation, food security, housing security, energy access, economic situation, education participation and attainment, community networks and participation, and health, happiness, and well-being. They will additionally investigate the question of whether the SMI reduces the administrative and bureaucratic responsibilities of social workers.
B-MINCOME is being administered by the Barcelona City Council, with consultation from five research organizations (the Young Foundation, the Institute of Governance and Public Policy at the Autonomous University of Barcelona, the Polytechnic University of Catalonia, the Catalan Institution for Evaluation of Public Policies, and NOVACT-International Institute for Non-Violent Action), and was awarded financial support by Urban Innovative Actions, an initiative of the European Commission formed to support projects investigating “innovative and creative solutions” in urban areas.
The leading political party in the City Council, the left-wing Barcelona en Comú, has expressed interest in implementing a municipal cash transfer program if results of the experiment prove favorable.
5. GiveDirectly’s Basic Income Experiment in Kenya
Status: Full experiment (consisting of 300 rural villages) planned to launch after the Kenyan election on October 26, 2017; pilot study running in one village since October 2016. [Update: Official launch occurred in November 2017.]
Official website: www.givedirectly.org/basic-income.
GiveDirectly, a US-based charitable organization, has been providing unconditional cash transfers to poor residents of Kenya and Uganda since 2009, when it was founded by a team of economists who had become interested in the hypothesis that cash transfers are the most effective means to combat extreme poverty. Since this time, its practice of delivering cash grants directly to those in need has proven efficient and effective, earning the organization recognition as one of GiveWell’s top charities.
In 2016, GiveDirectly announced a new ambition: the first long-term and large-scale randomized controlled trial (RCT) of universal basic income. The experiment will provide unconditional cash transfers to the residents of 120 villages, comprising more than 16,000 people in total, with some receiving payments for up to 12 years. This makes the experiment the largest of its type, in terms of the number of individuals who will receive assistance, as well as the longest in duration [4]. The cost of the experiment will amount to $3o million, most of which has been now raised by private donors.
In contrast to all experiments taking place in developed nations, GiveDirectly’s experiment examines the impact of programs that are indeed universal–that is, in which the unconditional cash transfers are distributed to all individuals residing in the villages where the program will be implemented [5].
Indeed, the experiment is designed as an RCT in which villages (not individuals or households) are the experimental units: 300 villages in rural Kenya will be randomly assigned to either the control group (comprising 100 villages), in which no cash transfers are given to any residents, or one of three treatment groups, in which all residents receive some form of unconditional cash transfer. In the first treatment group (comprising 40 villages), residents will receive cash payments of about 23 USD (€21)–roughly half of the average income in rural Kenya–every month for 12 years. In the second treatment group (80 villages), residents will receive monthly cash payments of the same amount, but only for two years. In the third treatment group, residents will receive a single lump-sum payment equal in amount to the two-year basic income (that is, about 276 USD).
As GiveDirectly explains on its website, “Comparing the first and second groups of villages will shed light on how important the guarantee of future transfers is for outcomes today (e.g. taking a risk like starting a business). The comparison between the second and third groups will let us understand how breaking up a given amount of money affects its impact.”
The organization further indicates that it will investigate outcomes including the following: “economic status (income, assets, standard of living), time use (work, education, leisure, community involvement), risk-taking (migrating, starting businesses), gender relations (especially female empowerment), [and] aspirations and outlook on life.”
GiveDirectly had initially planned to launch the experiment in September 2017, but has postponed the launch to after the Kenyan elections on October 26.
An initial pilot study commenced in one village in October 2016, in which all 95 residents now receive monthly unconditional cash payments, which will continue in this village for 12 years. This preliminary study is intended to help experiments fine-tune the implementation of the full experiment, and is not itself to be included in the analysis of the full experiment. Because of this, GiveDirectly is making public much of its data that is collects from the pilot village (e.g. responses to a survey of participants). (It will not, however, publicize data as it is collected from villages that are part of the experiment.)
GiveDirectly expects to publish its first results one or two years after the experiment’s commencement.
6. Y Combinator’s plans for a United States experiment
Status: Completed feasibility study in Oakland; issued draft of research plan for randomized experiment (3,000 participants in total) in two US states.
Official website: https://basicincome.ycr.org/
In January 2016, Silicon Valley entrepreneur Sam Altman (president of the start-up incubator Y Combinator) announced his intention to fund a test of basic income in the United States. Like many tech entrepreneurs, Altman cited concerns about job loss due to automation in explaining his interest in basic income: “I’m fairly confident that at some point in the future, as technology continues to eliminate traditional jobs and massive new wealth gets created, we’re going to see some version of this at a national scale.” The experiment is now the main project of the non-profit arm of Altman’s company, YC Research.
Since the initial announcement, YC Research has hired social work and political science PhD Elizabeth Rhodes as Research Director and assembled a team of expert advisors. This research team has designed and implemented a feasibility study in Oakland, California [6], and is now working to finalize the design of its full-scale experiment.
According to a project proposal released in September 2017, Y Combinator has decided to design the experiment as a randomized controlled trial, with a target population of low-income young adults–specifically, adults aged 21 to 40 whose incomes fall below the area median–in two US states. (Researchers will employ a stratified sample to ensure adequate representation across race, gender, and income categories.)
On the tentative design, researchers will select a total of 3,000 participants, of which 1,000 will be randomly assigned to the treatment group–whose members will receive unconditional cash payments of 1,000 USD per month–and the remaining 2,000 to the control group. (Individuals in the control group will provide the same type of feedback and data to researchers but receive only a much smaller cash payment, tentatively set at 50 USD per month, for their participation.) As currently planned, some individuals will receive the cash payments for three years, others for five.
Y Combinator emphasizes that its interest is a “holistic approach to understanding the individual-level effects of basic income”, in contrast to many past and present experiments which have focused on the labor market impacts of unconditional cash payments (such as Finland’s experiment and the negative income tax experiments conducted in the US in the 1970s). Among these individual-level effects, the research group is particularly interested in time use, mental and physical health, subjective well-being, financial health, decision making and attitudes toward risk, as well as political and social attitudes. Furthermore, although individual-level effects will be the focus of the experiment, researchers also hope to examine spillover effects on recipients’ families, friends, and communities.
While the research group has not finalized its choice of data sources and collection methods (see its project proposal for a discussion of possibilities currently under discussion), it plans to combine quantitative analysis with regular surveys and interviews. However, according to Rhodes, receipt of the cash payments will not be contingent on participation in surveys and interviews; payments will continue for the duration of the experiments even if recipients do not respond to requests for data and information.
To conduct the experiment, YC Research has partnered with the Center on Poverty and Inequality (CPI) at Stanford University. The research has been approved by Stanford’s Institutional Review Board for research involving human subjects. YC Research is also in the process of working with state and local governments to coordinate mechanisms for distributing payments without affecting recipients’ future eligibility for existing government benefits.
No specific launch date has been set for the experiment, as YC Research is still gathering feedback on its project proposal. However, at a plenary lecture at the 2017 BIEN Congress, Rhodes indicated that the research group hopes to commence the trial in early 2018.
7. Scotland’s plans for local experiments
Status: Government support pledged; preliminary reports in progress.
In November 2016, the Councils of Fife and Glasgow committed to investigate the feasibility of municipal-level basic income experiments. An important step forward occurred in February 2017, when the Glasgow City Council passed as a resolution to convene workshops on the financial, administrative, and constitutional feasibility of an experiment, partnering with the Royal Society of Arts (RSA), a British think tank responsible for a highly regarded report on potential for a basic income in the UK (“Creative Citizen, Creative State”).
Fife and Glasgow were later joined in their interest by North Ayrshire and Edinburgh.
In September 2017, the Scottish government announced its commitment to provide funding and support for basic income experiments in the four municipalities (see the RSA’s announcement), and RSA Scotland is now working with the government to prepare an initial report on the possibilities for an experiment in Glasgow.
No design decisions or prospective launch dates have yet been announced.
In general, the RSA and local authorities are attracted to basic income as a potential means to address poverty, precarious employment, economic insecurity, and the changing nature of work.
Notes
[1] The coverage of the present article is limited to projects that qualify as experiments in a social scientific sense. Thus, it does not include several projects that are sometimes listed–incorrectly–as “experiments” of basic income. Missing, for example, are lottery programs like Mein Grundeinkommen, as well as any pilot studies that lack a control group, such as the ReCitivas Institute’s pilot in Quatinga Velho and Eight’s two-year pilot study in Uganda (see “Some thoughts on basic income ‘experiments’” by Michael A Lewis).
Programs and policies such as Iran’s fuel subsidy reform and Alaska’s Permanent Fund Dividend are occasionally misconstrued in sentences like “Iran and Alaska have tested basic income”. These programs are also not experiments in the scientific sense, and thus omitted from inclusion in this article.
[2] Of course, no experiment can fully test the effects of a basic income, which, by definition, would guarantee cash stipends to all individuals in a community, and it would guarantee this support for life. Experiments, in contrast, are bound by their nature to be non-universal (since there must be a control group) and limited in duration. Furthermore, it is likely that additional changes to the taxation of benefit system, which are not captured in the experiments, would accompany the introduction of a basic income.
(In a forthcoming article, I will argue that, in the case of my own interest in basic income as a possible stimulus to long-term cultural change, the above facts preclude limited trials from having the capacity to produce usefully informative results.)
[3] According to many definitions, including those adopted by some of BIEN’s affiliates (although not BIEN itself), a “basic income” is a recurring payment sufficient to meet basic living expenses. On these definitions, it is indeed inaccurate to call describe the Finnish experiment a “basic income experiment” given €560 per month falls below average minimum monthly living expenses in Finland (although it might plausibly be called, as Kela sometimes more precisely describes it, a “partial basic income experiment”).
One may also raise concern regarding the universality of the program–or, rather, the lack thereof–given that a sample of adult recipients of unemployment benefits is clearly not representative of all Finns, and no “true” basic income (on any definition) would not be restricted only to the unemployed. One might alternatively say, however, that the Finnish experiment is indeed a test of a (partial) basic income, albeit one using a non-representative sample.
[4] In 2016, the Brazilian non-profit institute ReCivitas initiated a project, Basic Income Startup, which promises a “lifetime basic income” of 40 Reais (about 10 USD) per month to members of the village Quatinga Velho. The project targets volunteer recipients in areas where 40 Reais per month makes a significant impact on quality of life.
ReCivitas’s Basic Income Startup, however, is not an experiment (although occasionally miscategorized as one). Indeed, project founders have stated that they are already convinced that basic income is effective; their goal is not to test but to implement it.
[5] Previous trials of basic income and related policies have used saturation sites. Notably, Manitoba’s MINCOME experiments of the 1970s, which tested a type of minimum income guarantee in the form of a negative income tax, featured a saturation site in the town of the Dauphin.
The basic income experiment in Madhya Pradesh was also a saturation studym in which all members of the nine villages in the experimental group received monthly unconditional cash transfers (equal about one quarter of the median income in the state).
[6] Contrary to some misconceptions, the Oakland project was not itself an experiment. Its purpose was merely to test and fine-tune the mechanisms for conducting the experiment–such as the selection of participants, disbursement of funds, and collection of data–not to analyze the effects of unconditional cash transfers on recipients. The latter will be the goal of the project described in a newly released research proposal, and which has yet to be finalized and launched.
Addendum on Stockton, California (October 20, 2017)
Only a few days after this updated article was published, Michael Tubbs, Mayor of Stockton, California, announced his intention to carry out a pilot of basic income or guaranteed income in his city, the Stockton Economic Empowerment Demonstration (SEED).
The project will be privately funded, and has already received a $1 million SEED grant (if you will) from the Economic Security Project, a two-year initiative launched in December 2016 to fund basic income related projects in the United States.
At present, however, few details have been announced or decided.
According to its website, the project is “in the very early stages of development”.
Regarding the current state of the project, the site notes:
We will kick off the demonstration with a six-nine month design period that will prioritize community engagement and feedback. In that time period, we’ll identify research and storytelling partners […]. We will identify research priorities that complement existing research on unconditional cash transfers in the United States and invest in storytelling that honestly and authentically uplifts the experiences of recipients.
Thus, based on publicly available information, it is presently unclear when the project planned in Stockton will be launched, as well as what form it will take–including whether or not it will be an experiment (note the above emphasis on “storytelling” rather than comparison of data collected from control and treatment groups).
Basic Income News reporters will follow up with the conveners of the Stockton project, and provide more thorough, comprehensive and up-to-date information in a future stand-alone news article.
Reviewed by Heidi Karow
“Lab stuff” photo: CC BY-NC-SA 2.0 Rawbert|K|Photo
