Economy Lab

Should Canada have a guaranteed annual income?

The Globe and Mail

Many Canadians already effectively have a guaranteed annual income. Seniors, for example, are eligible for the Guaranteed Income Supplement. (Kevin Van Paassen/Kevin Van Paassen/The Globe and Mail)

Kevin Milligan is Associate Professor of Economics at the University of British Columbia



The idea of a guaranteed annual income (GAI) periodically surfaces in Canadian policy discussions as a transformational change to income support programs. Advocates can be found coming both from the left and the right. What is the GAI and should it be adopted?

Story continues below ad


The idea of a GAI is quite simple. Everyone receives a transfer from the government of some fixed amount. This transfer is 'clawed back' with every dollar of income received. This structure leads to benefits and income that look like the illustrative graph attached to this article, assuming a $10,000 initial transfer and a 25 per cent clawback rate.

People from the left find the GAI attractive because they like the idea of a secure and (potentially) generous benefit. People from the right like the idea of a lower tax rate and better work incentives for lower income people and a dismantling of the 'welfare wall'. Even highly stylized academic models show that optimal transfers might take this shape.



Ken Battle of the Caledon Institute stresses an important point about the GAI -- a lot of people already effectively have a GAI. Seniors are eligible for the Guaranteed Income Supplement, which pays about $8,000 per year but is clawed back by 50 cents per dollar of income. For families with children, an alphabet soup of children's benefits ( CCTB, NCB, WITB) follow the GAI structure, more or less.



Would it be worthwhile extending and enhancing a GAI to all Canadians? At least two major barriers stand in the way.



The first big problem faced by a GAI is politics -- federal politics to be precise. Some advocates of a broad GAI envision everything from public pensions to unemployment insurance to welfare programs to be rolled into one GAI program. If we were designing social policy de novo for a lunar colony that might make sense. But in the existing federal country of Canada, a GAI would require an utterly miraculous federal-provincial feat of co-operation. This is not to lament about querulous provincials holding us back from social policy bliss -- there are substantial differences in the types of needs across provinces that ought to be incorporated into policy and would be difficult to fit into a Canada-wide programme.



The second concern is cost. Even a more scaled-back GAI proposal that just replaced social assistance at a provincial level might still be very expensive. Having a transfer large enough to satisfy those worried about the wellbeing of the very poor, and also a clawback rate small enough to satisfy those concerned about work incentives might yield a very expensive policy. Targeted policies can hurt incentives, but universal policies are expensive: this is the great tension in policy design.



The support from thinkers of both the left and the right might portend broad social agreement on a GAI. I worry instead that the breadth of support is an indicator of a policy that hasn't been specified in enough detail, allowing the policy to be all things to all people. A good start toward a convincing argument for the GAI concept would be a fully specified and costed GAI that doesn't depend on miraculous feats of federalism to bring it into existence.