The basic income study that wasn't
When the British Columbia government announced in 2018 that it had appointed an expert panel to “test the feasibility of a basic income in B.C.”, there were high expectations among advocates that the results would be definitive.
While the report’s findings are well reasoned, they are not conclusive. Relying heavily on a particular conception of a “just society”, the authors produced a formidable set of calculations to make their case. However, the normative framework that they chose and the metrics that they prioritized led to a foregone conclusion.
An unconditional benefit, by definition, is an untargeted benefit. Hence an unconditional basic income can never improve on targeted outcomes when compared with conditional benefits that are, well, targeted—for the same amount of investment.
This is certainly the case with poverty reduction goals, on which the B.C. report concludes that a basic income approach would cost more to achieve than targeted measures. No surprise here.
The conclusions of the B.C. report are buttressed by what the authors claim to be an approach that is consistent with a “more just society”. They rely on a normative liberal, social welfare model of the just society inspired by political philosophers John Rawls and Elizabeth Anderson. This is a perfectly respectable theory, but it is premised on the primal importance (even duty) of paid work and on a strict view of reciprocity in social relations.
Relaxing either of those premises does not imply a less just society, only a different conception of what a just society means. The B.C. report prefers what might be described as “participation income” over “basic income” but this assumes that the economy offers dignified participation opportunities along the lines of what politicians like to call “decent middle-class jobs”.
The fact that this term even exists confirms that there are a lot of “non-decent” jobs in our market economy. The idea of “participation income” also values formal paid jobs over nonformal, unpaid forms of participation that may in fact offer greater personal reward, and social good—think of eldercare.
Surely the values of personal dignity and social benefit are as important to a “just society” as reciprocity?
The blinkers used in the B.C. report effectively turned a study that was supposed to be about the feasibility of a basic income in B.C. into one that was about designing a better income transfer scheme without using basic income. If the authors were serious about investigating the feasibility of basic income, they would have, at the very least, conducted a thought experiment in which feasibility could be tested under a different vision of the just society.
They did not have to agree with the alternative framework, but such an approach would have been faithful to what I believe was the point of the exercise.
Instead, they essentially brushed aside these factors by either claiming that the benefits were illusory or that there was insufficient data to calculate their worth.
It is true that there is little data to measure the broader and longer-term potential benefits of basic income, but on this I would make two points:
First, the authors did not have the same qualms in estimating the impact of transfer payments on the disincentive to work. These measures of “elasticity” are notoriously unreliable and extremely difficult to generalize across sectors, age groups, and regions.
What’s more, these estimates are usually based on short-run observations. There is every reason to suspect that long-run elasticities result in a lesser work disincentive, but the report is silent on this question.
Second, all feasibility studies face challenges of data unavailability or uncertainty, and the most sophisticated ones handle this problem by using tools such as break-even analysis.
For example: if it is plausible that there are social benefits from basic income in the form of cost savings to, say, the criminal justice system, what is the order of magnitude from such benefits that would make policymakers take a serious look at the idea?
This kind of reasoning does not appear in the report and it is because the framing of the study ruled it out from the outset.
One obvious way in which to “test” basic income is to conduct pilot projects. The authors reject this idea because they believe the short-term impacts of existing basic income-like programs are already known, and the longer-term impacts cannot be determined unless the pilot project is, by definition, long-term.
They then talk themselves out of a longer-term pilot project on the a priori belief that if basic income is not a good idea in principle, a long-term pilot project on basic income would be a bad idea in practice.
In fact, a pilot project could answer the central question that led the researchers to rule out basic income in the first place: how would the communities in which basic income was tested respond to unconditional transfers, given that it is purported to “violate” work obligations and social reciprocity essential to a “just society”?
Would the community respond in the way that the authors presume—that is, to consider these transfers “unjust”? For the answer to this question, I am inclined to look to the empirics rather than political theory.
As it turns out, that political theory is based very much on the thinking of Elizabeth Anderson, who is skeptical about basic income in theory, but open-minded about it in practice. It is unfortunate that the authors of the B.C. report do not share her open-mindedness:
“I am not persuaded that the costs to the goals of social democracy would be worth the gains provided by a UBI, but I am open to empirical evidence to the contrary.”
We await the empirical evidence.