Ontario’s Financial Accountability Office (FAO) recently released a report estimating that, on the assumption climate change will bring more extreme weather, it could add between $800 million and $1.5 billion annually to the cost of maintaining public buildings. To which Green MPP Mike Schreiner responded that there’s a “need to be honest with Ontarians about the costs of the climate crisis… the cost of inaction far exceeds the cost of climate action.”

If only! We should be so lucky if climate policy cost less than $800 million, or even $1.5 billion, annually. Even when we add up all the projected costs of climate change across the rest of society (assuming we could even agree on reasonable estimates), as has been noted many times, most climate policy is far costlier than the expected costs of climate change itself.

The contrast is even more stark when we do the comparison properly, by comparing the costs of policies, not to total damages but to the small subset of damages they would allegedly prevent. Which points to the flawed concept of “the costs of inaction.” It embeds a serious but subtle economic fallacy that, if left unchecked, vastly overstates the benefits of climate action.

The cost of inaction, by definition, is the difference between the costs incurred without the action and the costs incurred if the action is taken. If your mechanic says you need a $500 repair now to prevent a coming engine malfunction that will do $5,000 in damage, then the meaning is clear. The cost of inaction is $5,000, which is far higher than the cost of the action. But if the mechanic adds that the proposed repair won’t prevent the engine malfunction or mitigate the cost of the damage, the story suddenly changes. You face $5,000 in costs either way. Your choice is whether to pay an additional $500 in futile repair expenses. In this case, inaction is cheaper.

Same with climate. The problem with climate policy—specifically emission reductions rather than adaptation—is that the actions being talked about won’t change the outcome. The FAO used a set of global emission scenarios, not Ontario-specific emissions, because only the global total matters. Canada is responsible for about 1.9 per cent of global carbon dioxide emissions, and Ontario is responsible for about 22 per cent of Canada’s, making Ontario responsible for about four-tenths of one per cent of global emissions. People might have gotten the impression that emission reductions in Ontario would somehow prevent future Ontario extreme weather events. Nonsense. Even if we eliminated all Ontario emissions it wouldn’t make a difference, since what matters is the global total.

What makes the situation even more futile is that emission reductions in Ontario typically don’t result in global reductions. When climate policy raises the cost of doing business here (for instance, by our ill-fated green energy fiasco) the industrial activity doesn’t disappear, most of it just moves elsewhere, especially to Asia. We lose the investment and jobs but global emissions stay the same.

Even if they do go down a bit, it has long been known in climate policy circles that emission reduction treaties hardly affect the long-term outcome. In a 1998 study, climate modeller Tom Wigley showed that even full implementation of the Kyoto Protocol would have only slightly slowed down the accumulation of carbon dioxide in the atmosphere, delaying the date of carbon doubling after a century by a few years at most. Scaling down to Ontario’s 0.4 per cent share proves that our policy decisions are even more irrelevant. In a 2015 study, Bjorn Lomborg repeated the analysis for the Paris Treaty and found the same thing, for the same reasons. Despite the prohibitive costs of Kyoto and Paris, even full global compliance would barely change the global outcome, and Ontario-specific policies by implication have essentially zero effect.

Climate change may turn out to be costly, but that doesn’t mean the cost of inaction, properly measured, is more than zero. The benefits of climate policy are not given by the entire climate damage estimate, but by the reduction in expected damages attributable to the policy. This is basically zero regardless of how you do the measuring. And policymakers should start being honest with Ontarians about the cost of continually getting this concept wrong.