What would real job loss insurance look like?

Let’s suppose, for one mad, impetuous moment, that the coming bipartisan discussions on reforming employment insurance have something to do with reforming Employment Insurance. Bit of a stretch, I know, but just supposing. What would a reformed plan look like? What might the two parties come up with, if they were at all, you know, serious?

They might start by changing the name, or I should say unchanging it. It was Lloyd Axworthy who changed Unemployment Insurance to Employment Insurance, a piece of linguistic legerdemain that might be described as Orwellian if it weren’t so bloody obvious about it. You insure yourself in case something you hope will not happen does—hence fire insurance, flood insurance, accidental death and dismemberment insurance. In the same way, unemployment insurance pays benefits in the event of involuntary unemployment. Whereas employment insurance would seem intended to help you get past a temporary, unavoidable spell of employment.

Which, come to think of it, is a more or less accurate description of the system as it now stands. For a great many Canadians, Employment Insurance is not something they claim in the rare and unpredictable event that they lose their jobs. It is something they claim every year, as predictably as the changing of the seasons—exactly so, in what are known as the “seasonal industries.” It is as if the holder of a fire insurance policy were to claim for each of a calamitous series of annual house fires. And yet the system continues to pay out benefits, year after year, as generously as if they were first-time claimants.

Or rather, more generously: benefits are actually easier to obtain in regions where unemployment is relatively high—a plausible-sounding policy, were it not the case that these regions suffer such perennially high unemployment rates in part because of the easy availability of unemployment insurance. It’s pointless to debate whether those who claim benefits on such a persistent basis—as little as 420 hours of work, about 11 weeks, for as much as 37 weeks’ benefits—are malingerers, or merely responding to incentives. What’s inarguable is that the system effectively subsidizes employment in the seasonal industries, out of the premiums paid by industries with more stable employment records. Workers are encouraged to remain in industries where they are likely to become unemployed, not least because the system makes it pricier to hire them for anything else.

So any reform worthy of the name cannot restrict itself to meeting the short-term needs of workers caught in the grip of a deep recession. It should rather aim to put the insurance back into . . . whatever we wind up calling it. Yes, that means dispensing with regional variations in eligibility requirements, as everyone now seems to agree we should, in favour of one national standard. And yes, that cannot responsibly be achieved by bringing everyone down to the lowest standard of eligibility—still less reducing it to 360 hours, as the opposition proposes. Even capping minimum eligibility requirements at 560 hours in areas where unemployment is less than 10 per cent (currently these can range as high as 700 hours), as a recent TD Bank study has proposed, would cost another $500 million—though presumably this could be offset by raising the bar to the same level in other parts of the country.

But an insurance system should be based on insurance principles. At a minimum, that should include “experience-rating” of premiums, at least for the employer-paid portion, to eliminate the current cross-subsidization of one industry by another. Premiums in industries with a history of heavy use of the system would go up; premiums in more stable industries would go down. We would also do well to follow the C.D. Howe Institute’s suggestion, and set premiums with a view to balancing the books over the business cycle, rather than on an annual basis, as at present—a policy that, perversely, means raising premiums in bad times and lowering them in good, exaggerating swings in unemployment rather than tempering them.

Fine so far. But as much as unemployment insurance should be about insurance, it should also be about, well, unemployment. Over time, the system has been stretched to cover a number of other, ancillary concerns: first job training, then maternity benefits. These would be better funded out of general revenues: after all, why should job training be provided only to those eligible for unemployment insurance, and not to all workers? Is maternity leave really the sort of catastrophic event for which insurance was designed?

Either we move these social benefits out from under the umbrella of unemployment insurance or, inevitably, there will be calls to expand the system itself to cover those it now excludes. Indeed, that is happening now. Among the subjects the Conservatives and Liberals plan to discuss this summer is how to extend unemployment insurance to the self-employed—not just maternity benefits, as the Tories proposed in their last election platform, but basic unemployment benefits. How is this even possible? What will they have to do—lay themselves off?

Not that I expect such common sense from the journeyman pols appointed to the working group. That suggests an additional reform: take the politics out of it. Supposedly this was the point of the Tories’ new Canada Employment Insurance financing board, which will be responsible for setting premiums in the future. So why not quarantine the system from political interference more generally? Leave decisions about system design to an independent board of directors, with a single mandate: to make unemployment insurance about unemployment insurance.