William Watson: Paying people not to work could cost more than just money
Acting on a request from Conservative MP Pierre Poilievre, whose motives seem always ulterior, the Parliamentary Budget Office (PBO) reports that a national guaranteed basic income using the model Ontario is currently testing would cost almost $73 billion a year, with another $3.2 billion for an add-on of $500 a month for disabled people.
Because nobody really understands billions, another way to look at $76.2 billion is that it’s more than a quarter of Ottawa’s current program spending. It’s more than three times what Ottawa spends on interest every year. It’s almost four times the federal deficit, almost four times Employment Insurance benefits and almost four times defence spending. It’s a ton of money.
The way the current Ontario experiment works, if you’re single between 18 and 64 and you make no money at all, you get $16,989 — no strings attached. Why such a peculiarly exact number? Because it’s (arbitrarily) 75 per cent of Statistics Canada’s “low income measure,” which itself is (arbitrarily) 50 per cent below the median income.
From there any dollar you earn on your own reduces your entitlement by 50 cents. So if you manage to earn $33,978 — exactly twice $16,989 — the reduction in your $16,989 entitlement is $16,989 and you’re off the program. If you’re part of a couple and don’t have any earned income, you get $24,027, with subsequent earnings being “taxed” at the same 50 per cent. Basic income for a couple is less than basic income for a single person because two can live more cheaply than one.
How did the PBO figure out the program would cost $76 billion? (Note they actually break it down to $73 billion, with another $3.2 billion to provide an add-on of $500 a month for disabled people.) They looked at who’s making how much money now and, using a Statistics Canada simulation model calculated entitlements according to people’s earned income.
If you’re a single person currently making $10,000, you get $16,989 minus half of $10,000, which equals $11,989. If you’re making $20,000, you get $16,989 minus half of $20,000, which equals $6,989. And so on.
According to the model, as many as 7.7 million Canadians would receive money, with the average payout being $9,421. This is despite the fact that the number of low-income Canadians in 2014 was only 4.5 million. With the gradual phase-out of basic income, which makes sense for incentive reasons, some people will still get some basic income even if their total income means they aren’t officially “low-income.”
What’s wrong with this PBO estimate? Two known unknowns and a bunch of potential unknown unknowns.
First, will people still work as much if they’re getting an income guarantee? Probably not. That means more income has to be paid out than the PBO’s “static” estimates imply. How big is the decline in work? Nobody knows.
People have run simulations and there have been small-scale experiments from time to time and country to country. One simulation was co-authored by Jean-Yves Duclos, the current federal minister of families, children and social development, when he was an economist at Laval University. That study concluded: “Guaranteed income schemes may increase poverty rates and the incidence of low income rather than decrease them.”
Second, the federal government already spends lots of money on poor people. A PBO summary last year put total income-tested federal spending at $39.3 billion via 55 different programs, along with another $17.5 billion through 20 different tax expenditures. A basic income could, should and probably would replace some of these initiatives, which means its net cost would be less, maybe a lot less, than the gross cost cited above. Basic income could also replace programs and tax expenditures at the provincial and maybe the municipal level. But how many? And how much would that save? Every program has beneficiaries and supporters who won’t stop supporting what they may have worked hard to get.
A third set of unknowns is completely unknown. What long-term effects would come from a huge social change like guaranteeing people an income? Since our species began, income has had to be earned. If humans received a living income whether or not they work, what effect will that have on their attitudes to work, to each other, and to things we can’t imagine? Simulations using historical data and experiments embedded in small areas of what in this respect is still traditional society say nothing about that.
Fundamentally re-engineering a society involves bets so big they should make conservatives nervous, no matter what the simulations say.