What the 'inadvertent error' in the PBO's carbon tax analysis means, in as plain English as possible
Numbers in the 'counterfactual' scenario cited by carbon-tax critics will change: but by how much?
The next time you feel bad about a mistake in your line of work, spare a thought for the folks at the Parliamentary Budget Office.
One can only imagine the sinking feeling in their stomachs when they realized they had based their high-profile analysis of the federal fuel charge — commonly known as the carbon tax — on calculations that included more than just the federal fuel charge.
It turns out the PBO's complex computer code had actually included the federal output-based pricing system — commonly known as industrial carbon pricing — when it wasn't supposed to.
Whoops.
The PBO quietly updated a section of its website in mid-April to fess up to the mess up and retroactively added a note to its previously published reports. But it took until this week for many people to actually notice.
Speaking Wednesday on CBC's Power & Politics, Parliamentary Budget Officer Yves Giroux described it as an "inadvertent error," He said it would take until the fall for his office to rerun all its modelling and come up with corrected numbers.
Giroux also said he didn't believe the error would make a huge difference to the PBO's estimates of the "fiscal and economic" costs of the carbon tax. But University of Calgary economist Trevor Tombe isn't so sure about that.
"I think it would be very hard for anyone to know in advance what the results are going to be just based on gut feeling," he said.
So at this point you're probably wondering: If the PBO can't get this straight and other economists aren't sure what's going on, what hope do I have to understand all of this? It's a fair question. And the answer is: You've got this. Even non-experts can wrap their minds around a boiled-down version of Canada's carbon-pricing policies.
Yes, this is complex stuff. But it's also important stuff.
Canadian companies are making multibillion-dollar plans based on carbon pricing. The carbon tax is shaping up to be a primary issue in the next federal election, and climate change a defining issue of our global future. To say the stakes are high would be an understatement.
So let's break this down.
Wait. There are two carbon taxes?
When people talk about the federal "carbon tax," they are typically referring to the surcharge most Canadians pay when filling up their vehicles and the extra fee on their natural gas bills, which are accompanied by rebates that are direct-deposited into their bank accounts every three months.
This consumer-level carbon-pricing system is the target of Conservative Leader Pierre Poilievre's "Axe the Tax" mantra.
Poilievre has been relatively mum, however, on the other carbon-pricing system in this country: the one that large, industrial emitters are subject to. This form of carbon pricing long predates the consumer-level carbon tax. It even predates Prime Minister Justin Trudeau taking office.
In fact, it was Alberta that led the way on this front.
WATCH | At Issue: Will correcting the carbon tax analysis solve anything?
Way back in 2007, the oil-rich province became the first jurisdiction in North America to put a price on industrial carbon emissions.
Oilsands facilities and power plants and other large-scale industrial emitters are subject to this parallel system of carbon pricing and, instead of rebates, they receive "output-based" credits. In a nutshell, the more stuff they produce (whether that's oil, electricity or some other output) the more they are credited to offset their carbon costs.
This effectively rewards industrial facilities that can produce the same amount of stuff with fewer emissions.
For a high-efficiency natural gas power plant in Alberta, for example, it's a wash: their credits roughly cancel out the price they pay on their emissions. Less-efficient coal plants, by contrast, have to pay a significant amount. And emissions-free wind and solar plants actually make money from this carbon-pricing system. (All of which has helped contribute to Alberta's phase-out of coal and rapid adoption of renewable energy.)
The details vary from province to province — and even within provinces. (To make things even more complicated: Alberta's system is different for electricity generation than it is for oilsands!) Most provinces now have their own industrial carbon pricing systems that are compliant with federal law. Those that don't are subject to the federal Output-Based Pricing System, or OBPS for short.
And it's this OBPS that tripped up the PBO.
For years, they said they weren't including these industrial carbon-pricing systems in their estimates of the "economic and fiscal costs" of the consumer-level carbon tax.
But, as we now know, they actually were — by mistake.
So what are these 'economic and fiscal' costs?
The "fiscal" costs outlined in the PBO report are the literal dollars out of your pocket due to the carbon tax.
This includes both the direct costs you pay at the pump and on your natural gas bill, as well as indirect costs in the other (non-fuel) goods and services you buy — in the form of price increases due to companies passing on their own carbon-tax costs to the consumer.
The PBO isn't the only group to have analyzed these fiscal costs. A group of economists from the University of Calgary and University of Regina published their own estimates in a peer-reviewed journal and came up with almost the exact same numbers. There's little controversy, at least in academic circles, about these figures.
Both the PBO and the other group of economists agree the majority of Canadian households receive more in rebates than they pay in these direct and indirect fiscal costs, combined. (This is the portion of the PBO analysis that the federal Liberals like to point to, and that the federal Conservatives like to ignore.)
What's been more controversial, at least in the wonkish world of policy analysts, has been the PBO's additional estimates of "economic" costs.
This is a more complex exercise because unlike the fiscal costs, which exist in reality, the "economic" costs require imagining a "counterfactual" scenario in which the federal carbon tax doesn't exist — and no alternative climate policy replaces it.
The 'counterfactual' world
The PBO uses mathematical models to estimate what Canada's economic growth would be like in that hypothetical world, which is inevitably higher than it is in the real world with the carbon tax.
The difference between that "counterfactual" world and the actual world forms the basis of the "economic costs" of the tax.
Those economic costs, combined with the previously discussed fiscal costs, show a majority of Canadian households worse off with the carbon tax compared to the hypothetical scenario with no carbon tax and no other climate policy to replace it. (This is the portion of the PBO analysis that the federal Conservatives like to point to, and that the federal Liberals like to apply all sorts of asterisks to.)
Some critics have argued, however, that the PBO's choice of counterfactual is unrealistic. The non-partisan think tank Clean Prosperity has said it "models an implausible scenario in which carbon pricing exists in the absence of other climate policies, and in the absence of investment in decarbonization."
Economists, meanwhile, often argue that carbon-pricing, which leaves emissions reduction largely up to the free market, is the least costly kind of policy, especially when compared to more heavy-handed government regulation.
The Canadian Climate Institute has also said the PBO analysis "fails to consider economic benefits of carbon pricing and the costs of climate inaction, both in terms of stabilizing the climate and competing in a global economy racing to net-zero."
"Those broader factors are a huge part of the actual cost-benefit analysis around carbon pricing," it wrote in an article last year.
For its part as a neutral, non-partisan agent of Parliament, the PBO has said its report was never meant to be a cost-benefit analysis.
"We are mandated to provide estimates and analysis on the cost of policy proposals," Giroux said. "The government is usually quite adept at promoting the benefits of its own policies."
Giroux added that the PBO's choice of counterfactual scenario — a world with no carbon tax and nothing to replace it — is logical from an analytic perspective but isn't meant to be prescriptive.
"It's by no means intended to be an interpretation or a suggestion that doing nothing is the right thing to do," he said.
Again, though: the "counterfactual" scenario was originally meant to be — and presented as — one without the consumer-level carbon tax. But now we know the calculations included the industrial carbon tax as well, by mistake.
So what effect will that have on the PBO's estimates of the "economic" costs?
They'll be lower but it's hard for anyone to say, at this point, exactly how much lower.
Something for the PBO to 'reflect on'
As we've heard, Giroux doesn't think the "economic" costs will be all that different once the PBO is done rerunning all its mathematical models in several months' time.
But Tombe, who has done numerous carbon-pricing analyses of his own over the years, isn't so sure.
Including the industrial carbon price in the calculations "could have a large effect on the numbers," in his view.
"That pricing system potentially has a larger effect on future rates of GDP growth than the retail pricing system would," Tombe said.
Given the complexity of the modelling, Tombe said "the details really matter," but he believes the PBO's reports are often short on the kinds of methodological specifics that one would typically find in economic analyses published by other groups.
"Most of its reports are light on detail," Tombe said. "They almost never provide enough information for external analysts to even approach a replication exercise."
Tombe said the PBO is generally well respected but this makes its hard for outside economists to evaluate its work, which is something he hopes it will "reflect on" in light of this high-profile error.
"Because the details matter enormously in the larger-emitter system — how they're modelled, how they're simulated — I hope that they provide those details in whatever future report they come up with."
ABOUT THE AUTHOR
