The CMHC’s change to Mortgage Rules - A small lapse in logic
Last week the CMHC announced stricter underwriting policies for insured mortgages - a higher credit score of 680 is now required, debt service ratios have been reduced and non-traditional sources of down payment have been banned.
In normal times these changes are uncontroversial. Limiting the gross debt service ratio to 35% brings the metric (when combined with the stress test) in line with the affordability test typically used (30% of pre-tax income). The vast majority of mortgagors have credit scores above 680. And a down payment should be just that, a down payment, and not another loan.
In this article I am commenting on the direction of policy change right now which I think is not logically consistent with the CMHC’s mandate. I appreciate that the change is not necessarily significant and may have been driven by political considerations. The bigger thrust of my argument is really that there is significant public value in the government playing a larger role to ensure stable real estate values.
Consider the following 3 statements:
An important part of the CMHC’s mandate is to contribute to the well-being of the housing sector in the national economy and the stability of the financial system.
Recently the CMHC predicted that Canadian real estate values could drop by as much as 9-18% over the next 12 months.
Last week the CMHC announced stricter underwriting policies. Stricter underwriting for insured mortgages will reduce demand and, all else being equal, lead to lower real estate values.
Any 2 of the above statements are logically consistent, but not all 3.
For example, if we ignore #2 (or replace it with a concern that real estate values will increase due to the large injection of liquidity and ultra low rates occasioned by Covid-19), it would be logically consistent to tighten credit to reduce Canadians’ debt levels and ensure real estate values don’t accelerate further, therefore contributing to financial stability.
If we ignore #1 and think of the CMHC as a pure commercial entity, we could argue that given the inherent risk in the system the CMHC wants to reduce its exposure to Canadian real estate and is simply using the credit lever to achieve that.
To be clear, applying underwriting criteria with more discipline is part of managing a lending or insurance business in a downturn. I would imagine mortgage insurers and banks are spending more time ensuring the sustainability of income right now. Given the rise in unemployment, mortgage origination volumes will decline. The fact is not always appreciated that even with the same underwriting criteria a bank originates better quality credit in a downturn. Put simply, people who still have jobs even in a downturn are better credit. Or a credit score of 680 at the peak of the cycle counts less than the same score in the trough.
During the steady rise in real estate values over the last decade, the Department of Finance, OSFI and the CMHC made several changes to policy to rein in the market e.g. by reducing amortization, implementing stricter requirements for insured mortgages, increasing mortgage rates indirectly by limiting the volume of CMBs and introducing stress tests. These all made sense during an economic expansion with rising real estate values.
Some of these changes were justified by saying that the CMHC was reducing taxpayers’ exposure to Canadian real estate. While I agreed with the changes, I don’t agree with the justification. The government (and taxpayers) always implicitly carries real estate risk. In Canada the government actually gets paid for this risk through premiums paid by Canadian homeowners to the CMHC. The federal government benefits from an annual dividend in the $2B range from the CMHC.
Why does the government always implicitly carry real estate risk?
When a real estate bubble bursts, people lose their homes, and/or a significant part of their savings. The federal government recognizes that adequate housing is a fundamental human right and therefore has an obligation to provide housing for Canadians who cannot afford to do so themselves. As we saw during the Financial Crisis of 2008 in the US, when a real estate bubble bursts it puts banks at risk. Again, the government would have to step in to save the banking system. Finally, in a democracy there is a strong incentive for the government to ensure the majority of citizens are financially secure. A home is the biggest investment for most people.
For those who fret over the fact that banks do not carry the risk for insured mortgages but earn the reward, the answer is simple: If the government suffers a significant loss due to mortgage insurance, nothing would stop it from taking back the loss from the banks by simply taxing them. It’s like re-insurance. You make a claim today, I increase your premiums (or taxes) tomorrow.
If it turns out that Canada is heading towards a deep recession and, among other things, real estate prices were to decline by 9-18%, the Department of Finance, OSFI and the CMHC should do the opposite of what they did in the last decade. They should gradually make more mortgage insurance available, increase amortizations, increase the availability of CMBs and reduce the stress test.
But why not be clearer on what we expect? Why not set a target range for real estate value growth (e.g. -4% to +6%) similar to the inflation target for the Bank of Canada? Imagine the social welfare that could be created by stable real estate values and the random reallocation of wealth that would be avoided in the absence of real estate bubbles.
The corollary to this is that regulation would have to be regional. Toronto may be going up and Calgary down. OSFI has to be willing to mandate a different stress test in Toronto than in Calgary.
It can be argued that manipulating real estate values through the availability of credit is scrambling the economic signal that would stimulate an increase in supply when required. However, the economic signal is already scrambled by policy and is severely delayed. Greenfield development in Toronto takes more than a decade. It’s like sending a WhatsApp message to your cousin on Alpha Centauri: by the time she gets the message it’s old news; when she responds it’s ancient history.
Now wouldn’t it be nice if we had one person who was responsible for stable real estate values in Canada?
A Tiff Macklem for Housing. Of course, this person would have to control all the significant policy levers for housing finance, which are now under the control of more than one ministry.
CMHC’s comments on Covid-19 in their 2019 annual report (published in early May) is reassuring: “According to our Risk Appetite Framework, our role as a key stabilizing component of the Canadian financial system means we will be substantially increasing our appetite for risk as we and other institutions absorb the impacts of these events.”
The recent tightening of underwriting policies seems like a small lapse in logic.