A recent innovation for political campaigns is that they can ask the Parliamentary Budget Office to provide an estimate for the impact of any taxation proposals, and NDP leader Jagmeet Singh was the first out of the gate to have his “super-wealth tax” evaluated. The PBO estimates that the levy would net out some $6 billion in the first full year of implementation. Sounds like a lot of money! Colby Cosh explains why it’s not quite what it might seem:
Federal NDP leader Jagmeet Singh taking part in a Pride Parade in June 2017 (during the leadership campaign).
Photo via Wikimedia.
Alas, the bean-counters always swoop in to spoil things. Singh’s wealth-tax scheme is instructive not only because he availed himself of PBO costing, but because it usefully reveals the limits of what the PBO or any other economic modeller can do. Look, in other words, at the fine print.
The PBO’s job was to estimate what you can extract from “an annual net wealth tax on Canadian resident economic families equal to one per cent of net wealth above $20 million.” In the PBO model this is a simple multiplication, but the roughly $6 billion take is arrived at only by reducing the revenue by 35 per cent to correct for “behavioural response” — that is, lawful (and unlawful) tricks employed to avoid the new tax by the rich targets. The net revenue is what’s left after you deduct another two per cent to cover administrative costs.
And, as the PBO immediately insists, “the estimate has high uncertainty” on both counts. This means they’re educated guesses. Jennifer Robson, a social policy prof at Carleton University’s Arthur Kroeger College, pointed out on Twitter that right now we don’t tax economic families per se and we don’t report assets and debts routinely to Revenue Canada. Ideas for pure wealth taxation (which is rare in practice) are predicated on the creation of, essentially, a new tax system — one which would have to detect and perpetually update how much, for example, the furniture in your house costs. The 35 per cent loss from behavioural response is at the high end of historic estimates from real-world examples. Even within our current tax system, Robson observes, we only get two extra dollars for every one we spend on expanding collections and compliance against the existing tax base.
As a practical matter, a wealth tax would mostly be, or would act most efficiently as, a tax on bank balances and investment accounts. Of course, there is always real estate. The super-rich seem to have a lot of that, and it is relatively easy to tax, and the resentment of Torontonians and Vancouverites who don’t own some is, for better or worse, a major reason the NDP is trying to weaponize envy.
But this reminds us that property taxes and taxes on property transfers perform a similar function, although they are not used primarily for income redistribution as such here — and in Canada ours are relatively high. The OECD does a little league table of tax structures, and compared with other industrialized countries Canada’s take from property taxes is about double the average. In a 36-country list we are near the bottom (33rd) in our dependence on taxing goods and services, and about average (12th) in dependence on corporate taxation, but fifth highest in dependence on personal taxation — and third in dependence on property taxes.
