Thorsten Koeppl is associate professor at Queen's University and scholar in financial services and monetary policy, C.D. Howe Institute. James MacGee is associate professor of economics, University of Western Ontario, and a research fellow at the C.D. Howe Institute.
The rise in house prices – especially in Toronto and Vancouver – has coincided with a build-up of risks in the mortgage market and homeowners taking on too much debt relative to their income.
In response, the federal Department of Finance has proposed shifting some of the risk of mortgage defaults onto lenders from insurers, through a mortgage insurance deductible that would come out of the lender's pocket in the case of mortgage default. Good intent. Bad idea.
The hope is that by shifting some of the potential losses from mortgage defaults onto lenders, a deductible would reduce the number of risky mortgage loans originated by lenders. But introducing a mortgage insurance deductible is a blunt and ineffective tool that would do little to reduce risk in the housing sector. Even worse, besides increasing administrative costs for mortgage insurance, it would undermine the role of Canada's mortgage insurance system in stabilizing the Canadian economy should a housing crash occur.
Insurance plays a central role in Canadian housing finance. Federally regulated financial intermediaries cannot offer high loan-to-value (80 per cent or higher) mortgages without mortgage insurance. In addition, some lenders encourage borrowers to take on mortgage insurance so as to be able to access CMHC-run securitization programs. As a result, roughly half of all mortgages in Canada are insured.
Proponents of a mortgage insurance deductible argue that shifting some potential losses back to mortgage lenders offers two benefits for Canadians. First, as the Department of Finance argues in a recent consultation paper, a deductible would address moral-hazard concerns. These concerns arise since lenders have an incentive to extend credit to high-risk borrowers if they can shift all of the default risk to mortgage insurers. This would help discourage the build-up of risky mortgages. Second, this shifting of risk would reduce the potential exposure of taxpayers to federally insured mortgages in the case of a crash.
These arguments don't hold up to closer scrutiny. The reason that the government guarantees mortgage insurers is to prevent a housing-crash scenario – should one occur – from morphing into a financial crisis. By back-stopping mortgage insurers, the federal government reduces the risk of Canadian financial institutions being hit by large losses, which removes a potential trigger for a financial crisis. To maintain the effectiveness of our mortgage insurance system as insurance against a full-blown financial crisis, deductibles would need to be capped at a relatively modest level, to avoid threatening the solvency of lenders in a housing-crash scenario. Once passed on to borrowers, however, such modest values for the deductible imply small increases in mortgage rates for riskier borrowers. This implies that a deductible would do little to discourage lending to borrowers who are at relatively high risk of defaulting on their mortgages.
A better way of dealing with excessive mortgage lending to high-risk borrowers is to combine a change in the pricing of mortgage insurance with continued refinement of the regulation of the mortgage insurance system. First, risk-based pricing is needed. Currently, mortgage insurance premiums do not take into account how default risk differs across mortgages, and borrowers, beyond the loan-to-value ratio. Charging the lender an insurance premium that takes into account risks associated with specific characteristics of the borrower would directly address moral-hazard concerns.
Second, this shift to risk-based pricing should be complemented with targeted reform of mortgage insurance underwriting regulations. This is especially important given growing evidence that the tightening of mortgage insurance rules has led to a rising number of high debt-to-income mortgages that are not covered by mortgage insurance and are originated through non-standard lending channels. Policymakers should rely on other tools such as interest-rate stress tests under which lenders are required to assess whether uninsured mortgages could handle higher interest rates to discourage this channel for growing risk.
Although reforms of mortgage insurance and mortgage underwriting since 2010 have slowed the build-up of vulnerable borrowers, more is needed to contain risks in a rising interest-rate environment. This means that targeted and effective policies are needed. A deductible on mortgage insurance as floated by the Department of Finance, however, does not meet the test.