Edmonton and Calgary minimum income pilots would be messy, but should be tried anyway

Guest blog by Carter Vance

One of the more curious, and encouraging, effects of the Alberta NDP's recent electoral victory has been the seemingly uncorked enthusiasm in the province for new-thinking legislative initiatives, from the hard limits on corporate and union donations to the open floating of a carbon tax as an emissions-reduction solution. It's a classic version of cut-the-knot politics, where roadblocks previously caused by a wound-together set of interests from a longstanding political dynasty and its intimate enmeshment with a variety of stakeholders and lobby groups are now much more easily overcome due to the presence of relative outsiders.

Though Rachel Notley and Co. have gone to great lengths to reassure businesses and moderate voters that they're no radicals, and indeed the rhetorical wet kiss Ms. Notley gave to the oil industry in a recent speech struck at least some progressives as a bridge too far in this direction, they are nonetheless showing a willingness to think outside the box. Whatever one thinks of these policies, and those yet to come, it can scarcely be denied that Alberta seems the place where fresh policy thinking is most likely to get an open ear from government decision makers.

No more so has this been clear than in the enthusiasm shown by finance minister Joe Ceci for the notion of a guaranteed minimum income (GMI) program. Calgary and Edmonton Mayors Naheed Nenshi and Don Iveson have offered their cities as test cases for implementation, each showing a wonkish spunk for the idea belying the conventional wisdom that such a program would never go over with voters. Ceci should reward this interest by appropriating the necessary funding for city-based pilots in the upcoming fall budget. The main barrier to widespread implementation of a GMI, or even the investigation of its potential viability, is a lack of available test cases, especially on a large scale. Anything that tests the viability of such a proposition is to be welcomed, whatever the results might be.

To briefly describe, the idea of a GMI is quite simple, though the exact form it might take can greatly vary depending on exact design criteria. The government would, in essence, guarantee that all citizens had access to a certain minimum standard of income, usually set just above the poverty line, either via direct payments (i.e. everyone gets a cheque worth the poverty-level income) or via a negative income tax (i.e. individuals with income below a given threshold get supplemental pay from the government rather than paying taxes).

The money to pay for this would come from eliminating or scaling back other forms of income security payments (from EI to public pension schemes to traditional welfare), and their related bureaucracies, though where to draw the line with regard to should be considered phased out in favour of a potential GMI is a matter of some controversy. More doctrinaire advocates suggest that minimum wage laws and at least some economic development projects could be eliminated if a GMI were to be implemented, while more minimalist GMI proponents would prefer to see only “safety net” payments (e.g. traditional welfare, but not pay-in programs like EI or the CPP) eliminated.

Ultimately, whatever form a GMI might take, we shouldn't kid ourselves: it wouldn't be cheap. A back-of-the-envelope calculation of multiplying the most recent Low-Income Cut-Off (LICO), a popular poverty line, for single individuals by the total population reveals a price tag of $679 billion. Now, this is an obvious overestimation, but it isn't chump change, to say the least. That is also taking the most directly expensive form of a GMI (the “cut-everybody-a-cheque” route).

A more likely route to a GMI, the “negative income tax” approach, would be significantly cheaper, at least as a headline figure. A quick calculation – based on the average before-tax LICO (about $28,000) and the average poverty gap ratio (35.7%) – suggests that, on average, low-income families would need about $10,000 to bring their incomes to the poverty line. Multiplying this figure by an estimate for the number of low-income families according to the before-tax LICO (about 2 million), we get $20 billion as a rough estimate of the total cost of the program. It should be noted that this estimation assumes a 100% clawback of benefits as income rises for people below the poverty line. In addition, it does not count for potential lost revenues and work disincentives resulting from the introduction of the program, i.e. some may work less or stop working altogether because of the program, thus raising costs. To reduce these disincentive effects, designers of the negative income tax could lower the clawback rate, but this would raise the total cost of the program.

Though it is difficult by nature to find an exact comparison in current expenditure terms for this (as per the aforementioned debate on just what programs should be eliminated in favour of a GMI), the latest numbers from the OECD indicate that Canada spends approximately $319 billion per year on “social expenditures”, which the OECD defines as separate from health and education, and therefore mainly encompassing direct cash payments and labour market activation program of various sorts. It should be noted that the exact figure comparisons between a basic income and the status quo will be highly variable depending on the program’s exact design. As well, such cost estimations do not take into account potential savings in other areas, most notably health care, which could be reasonably predicted to occur from a basic income provision.

There are, of course, the objections that not all of the current welfare bureaucracy is useless overhead which could be easily chopped, and certainly there is truth to this. Many support programs covered under that “social expenditure” category are genuinely useful to clients, simply giving money may not be adequate to cover in-kind needs that some individuals have and there are of course the genuinely hard cases of persons, for any number of reasons, not able to adequately manage their own funds who would be made worse off and perhaps more vulnerable to unscrupulous actors by such a move. Whilst that ought to sound a note of caution and give reflection on how to best design safeguards in the event of a GMI being implemented, it is also true that current welfare bureaucracies are often deliberately invasive and humiliating for people seeking assistance, they can create perverse incentives for beneficiaries improving their situation (for example, the loss of drug and dental benefits when moving off of Ontario Works), and they spend gobs of money hiring staff whose basic job is to sift for any evidence of misappropriated funds, which likely would've cost far less to simply have leak than to hire the those staff. The basic logic that people generally understand what is best for themselves, and simply need adequate resources to be able to provide those things, undergirds the case for a GMI; it is as sturdy a principle as any in public policy and deserves a serious attempt at implementation.

The idea of a GMI has been floated by thinkers economic both left (who like its guaranteed standard of living and the idea of cutting the link between work obligation and basic necessities) and right (who like the idea of gutting the welfare bureaucracy and minimizing the role of the state in service delivery), but hasn't ever truly caught on in terms of actual government policy.

Canada does currently practice a form of a GMI for senior citizens via the GIS top-up to the OAS for seniors without other sources of income (this system most resembles the negative income tax form of the GMI), and, as all policy junkies will recall, there was the until-recently-forgotten “Mincome” experiment in Dauphin, Manitoba from 1974 to 1979, which found a variety of effects that shouldn't be too surprising. Work participation fell only amongst young persons (who stayed in school longer) and new mothers (who took more time off to raise their children), whilst public health indicators improved a great deal, as did stress levels in the community and, by definition, poverty in essence disappeared.

Other studies done in the US around the same time returned a more mixed set of results, which did find higher disincentives to work in areas where a GMI was implemented, though it should be noted that this mainly came in the form of people taking longer periods of unemployment in order to find new jobs more correctly aligning with previous skills and credentials. This concern about diminished work incentives forms the standard case against the GMI and it should not necessarily be dismissed out of hand. It is true that labour market activation should be an overall goal of government policy, though it is questionable whether it makes a great deal of sense to prod an extra two years of labour out of a 63 year-old manual worker with the threat of destitution. It is also true that, as more and more job tasks become automated, there may simply be a lot less work for people to do, and, in the interest of avoiding complete social bifurcation, the case for a GIM becomes all the more necessary.

There are doubtless some who, in the event of a guarantee of minimal living standards, would be content with this and simply drop out of the labour force entirely, but I greatly suspect that won't be the case for most people. Quite beyond the monetary benefits, the jobs people actually enjoy having tend to be done for a variety of more intangible reasons (e.g. opportunities to build social networks, feelings of personal satisfaction, the work itself being interesting, etc.), which wouldn't change with a GMI. As to the more unpleasant jobs that are nevertheless necessary for society to function properly (e.g. garbage collection), I would simply suggest that perhaps they would need to pay more or offer greater benefits to incentivize workers toward them. There may be other effects, both benefits and drawbacks, to a GMI, but the fact remains that until we get more recent and relevant data on this we simply won't know. The Dutch city of Utrecht appears to be moving ahead with a pilot of similar type to that floated by Iveson and Nenshi, and its effects will certainly be interesting to observe However, since the test is situated in a much a different context in terms of economics and welfare state design, any conclusions will be less relevant to the Canadian context.

As for how this might work on such a small scale, in Edmonton and Calgary, here is where things start to get a bit more messy. Presumably, if this pilot were to go forward, it would need to be based on the distribution of minimum income benefits to residents of both cities. The problem here is that there's little in the way of a hard-and-fast manner of determining who qualifies as a “resident” for those purposes. One supposes the cities would use registered addresses, but then how would this affect people who move in or out of the city after the start of the pilot? Are they simply out of luck? What about persons with no fixed address? What about students, persons who reside in the cities seasonally, or people who work in the city but reside outside of its limits? Further, as social protection, broadly speaking, is embedded in the provincial jurisdiction (in terms of the “last resort” income support programs, at least), how would these pilots interact with the rest of Alberta's welfare system? Would persons in Edmonton and Calgary already receiving income support or disability payments be shifted onto the minimum income program? Receive additional payments?

This line of thought can go on, but it makes clear that the matter and potential of the GMI as a method for cutting through the tangled mess of program overlap which characterizes the current welfare state needs to be taken on at a higher and more coordinated level in order to be properly assessed. Whatever is revealed by these pilots, if and when they occur, is likely to be mostly in terms of labour market effects and social condition indicators. These findings will be nothing to scoff at, and may indeed persuade skeptical policymakers to go further, but they may tell us less than we want to know about what a GMI can and can't do.

That being said, Iveson and Nenshi should be applauded for bringing this issue to the public fore, and they, along with the new government of their province, should be encouraged to pursue, design and implement the basic income pilot projects in their cities.

Canada is desperately in need of genuinely new thinking on any number of policy areas, those of poverty reduction and the condition and design of the welfare state not the least amongst them, and such a pilot would provide a way forward. Basic income is a beloved idea of the policy wonk class, but it suffers from a lack of concrete, workable designs to convince a no-doubt skeptical public that it can in fact work in a sustainable way. Anything that moves this conversation forward, and indeed it may be found that the program simply does not work in reality, should be welcomed with open arms. Ironic though it may be, what was often joked of as that most hidebound of provinces may be able to illuminate the key decisions on the future of social services for us all.