Bank of America Merill Lynch has released a new report about Canadian real estate titled, “Our Homes Have Four Walls“. Hat tip to Dylan for emailing this to me.
The main conclusion of the report is as follows:
“Valuation metrics of Canadian housing are clearly stretched, but the usual
symptoms of a tipping point are simply not there. Speculation is low, price
expectations are cautious, home building is not excessive, and most importantly,
the economy continues to expand steadily.”
Now let’s see how they’ve come to that conclusion and if they’ve missed anything along the way.
On the difference between the Canadian and US mortgage market
“Focusing on the key differences between the Canadian and US mortgage market,
we find that the structure of the Canadian mortgage market greatly reduces the
probability of a US style housing melt-down. In the US excessive risky lending left
financial institutions vulnerable, a housing market vastly over supplied and
subsequently created a record breaking foreclosure crisis”
Nothing to argue with here. Even the most ardent housing bear has to acknowledge that indeed there were significant lending differences between the US and Canada. But to acknowledge difference does not dismiss the reality that we have had our own “made in Canada” lending excesses characterized by dramatic loosening of mortgage lending standards. So while we gaze south of the border at the greatest wealth destruction in human history, let’s not get entirely complacent about the prospects that we may have engaged in our own excesses, though arguably on a smaller scale.
“We find government guaranteed mortgage insurance mitigates risk to
financial institutions. Unlike the US where financial institutions were
clearly over exposed and the solvency of insurance providers were
Indeed this has mitigated risks to our for-profit financial institutions and has instead placed the risk squarely on the shoulders of the taxpayers via CMHC. While the big Canadian banks have seen a net reduction in residential mortgage exposure over the past 3 years, they nevertheless have significant exposure to non-secured debt such as unsecured lines of credit and credit cards. The exact extent is unknown, but it is worth noting that a typical pattern of default involves skipping payments on unsecured debts before missing a mortgage payment.
Legal recourse laws reduce the risk of households walking away from
their mortgage and implicitly improve lending quality, unlike the US
where reports of abandoned vacant homes were and remain rampant.
This remains one of the most misconstrued facts. It still surprises me how many people miss the fact that over half of the US states are in fact recourse states, including some of the hardest hit such as Florida, Nevada, and Michigan.
In fact, states that have an element of lender recourse outnumber those that are non-recourse.
This is not to suggest that having an element of recourse has no impact on default rate. In fact, the same Fed paper referenced above also suggests that recourse reduces defaults by 20%. Not insignificant, but it certainly does not reduce them completely as seems to be often suggested, and it no doubt provides little comfort to those living in some recourse states like Florida and Nevada where the delinquency rate is hovering around an astonishing 20%.
Government backing of 30% of the mortgage funding market acts to prevent US style funding freeze which would amplify the problems.
This fact is true until all of a sudden it’s not. While theoretically it should serve to protect against a funding freeze, we can also point to the fact that the Canadian government and the Bank of Canada had to take extraordinary measures in 2008 and 2009 to prevent the very funding freeze mentioned above.
On the conservative Canadian consumer
“Canadian’s have historically held lower leverage ratios than their US counter parts and tend to gravitate to more conservative mortgage options. Canadian household balance sheets have deteriorated and have been treading into more risky areas like variable rate mortgages, but sub prime lending remains a virtually non-existent market in Canada.”
Let’s dissect this statement one point at a time. I suppose that the statement that Canadians have ‘historically’ held lower leverage ratios is entirely accurate, if not misleading. The more important question is what do these leverage ratios look like today, and what do they look like in a historical context.
If we were to take ‘leverage ratio’ to indicate total debt levels relative to incomes, we would note that Canadians are at leveraged at a historically high level. By some measures, our current debt levels exceed those of the American consumer before their crisis. By all measures, we’re currently more indebted than they are.
If perhaps the statement is specifically directed at the level of real estate leverage, it may be worth revisiting the average level of house equity in Canada. You’ll recall that while Canadians have an average of 63% equity in their homes, the US had 60% equity in theirs before their bust. This is to say that for every dollar of home value, the average Canadian has 63 cents in equity and 37 cents in debt. This is not remarkably different from that US experience and it highlights just how much a rapidly rising real estate market can mask an expansion in debt as the leverage ratio still looks relatively strong.
With regards to gravitating towards more conservative mortgage products, there is nothing more conservative than a 30 year mortgage term in which the interest rate is fixed at a ridiculously low 4-5% for the entire length of the mortgage. This is the mortgage product that dominates the US market. Contrast that with a market where the vast majority of the population is exposed to interest rate risk every 5 years. This is the Canadian market. Consider that the Bank of Canada recently ran a stress test scenario in which they determined it would take only a 0.5% increase in interest rates for 1.1 million Canadian households to become at risk of defaulting on their consumer credit or mortgage-related debt. Now how conservative do we look?
Finally, the report states that sub prime lending is virtually non-existant in Canada. I suppose it’s important to define that term before we discuss whether or not it is happening in Canada. While a precise definition is not set in stone, “As a rule of thumb, a subprime mortgage is a home loan to someone with a credit score below 620. But some lenders count loans as subprime even if the borrowers have credit scores of 660 or higher, if the borrower makes a down payment of less than 5 percent or does not document income or assets.“- Source
So does this happen in Canada? According to CMHC, a borrower needs only a FICO score of 600 to qualify for mortgage insurance of up to 95% of the value of the home if the mortgage is a fixed rate product. This rises to a whopping 610 if it is a variable product.
Consider also the fact that Canada still has zero down mortgages. While structured as 5% cash back products, they nevertheless serve the same purpose of allowing homeowners to purchase a home with zero equity as they can borrow the necessary 5% down payment and then pay it off with the cash back when the sale closes, leaving them with zero equity.
We also have stated income mortgages where self-employed individuals don’t have to prove their income.
Taken collectively, these facts certainly cast doubt on the statement that sub prime lending is ‘virtually non-existent’ in Canada. While perhaps not structured as the ARMs and Alt-As that crashed the US mortgage market, they represent lending standards that are far from conservative.
I have to leave it here for now. I’ll make another post later today where I’ll deal with the rest of the inaccuracies in this report as well as hopefully look at the latest CREA stats.