NOVEMBER 06, 2012
Yesterday I began a short series commenting on an excellent CIBC report. In part 1 of this series, I gave credit where it was due. Today, I'll outline some disagreements with the report's central assertions. Part 3, dealing with additional considerations, can be read here.
Where I disagree:
1) Prevalence of subprime lending in US and Canada:
As I said above, Canada simply did not copy some of the stupidity seen State-side. That said, I believe the difference has been overstated. For starters, seldom are the terms “subprime” and “non-conforming” fully defined in these discussions. I would think that an article that makes its central point around the claim that “at its core, the US meltdown is a non-conforming story” would want to adequately define that term.
Canadians tend to think of a “subprime” borrower as someone with no income, no job, no assets, and no real hope of ever repaying the loan. While that subset of subprime loans were particularly toxic, it would be a mistake to assume that it is indicative of all non-conforming loans.
For insight into this topic, let’s turn to a Fed Reserve of Cleveland 2009 research note titled "Ten myths about subprime lending":
Subprime mortgages went to all kinds of borrowers, not only to those with impaired credit. A loan can be labeled subprime not only because of the characteristics of the borrower it was originated for, but also because of the type of lender that originated it, features of the mortgage product itself, or how it was securitized.
Specifically, if a loan was given to a borrower with a low credit score or a history of delinquency or bankruptcy, lenders would most likely label it subprime. But mortgages could also be labeled subprime if they were originated by a lender specializing in high-cost loans—although not all high-cost loans are subprime. Also, unusual types of mortgages generally not available in the prime market, such as “2/28 hybrids,” which switch to an adjustable interest rate after only two years of a fixed rate, would be labeled subprime even if they were given to borrowers with credit scores that were sufficiently high to qualify for prime mortgage loans.
[…] The myth that subprime loans went only to those with bad credit arises from overlooking the complexity of the subprime mortgage market and the fact that subprime mortgages are defined in a number of ways—not just by the credit quality of borrowers. One of the myth’s byproducts is that examples of borrowers with good credit and subprime loans have been seen as evidence of foul play, generating accusations that such borrowers must have been steered unfairly and sometimes fraudulently into the subprime market.
In some ways it’s no different than some of the arguments we hear for allowing longer amortization periods on insured mortgages and fewer regulations on HELOCs here in Canada: It allows some people to divert their capital to more profitable ventures such as their business of other investments.
In the same way, some “subprime” borrowers could have qualified for prime lending but chose not to for a number of reasons.
Also of note, that same Cleveland Fed research note pegged the “subprime” market at 16% of all outstanding mortgages in 2008, significantly lower than the 20% in 2006 quoted in the Tal piece, though some of the difference could have been because some of these mortgages would have already defaulted and been written off the books.
“Subprime” and “Non-conforming” never defined in these discussions:
Since the Tal article chose not to define some very key terms, I’ll take the opportunity to add some context.
The reality is that there are conflicting definitions of what “subprime” really is, and it was labeled in a number of ways in the US. This was discussed in a previous post, but the bottom line is that two of the common ways in which subprime loans were defined are based on credit scores and loan-to-value at origination.
In a comprehensive analysis of the precursors of the US housing bust titled, ‘Making Sense of the Subprime Crisis’ published in Brookings Papers on Economic Activity, the authors noted the following:
“We argue that the increased number of subprime loans that were originated with high loan-to-value (LTV) ratios was the most important observable risk factor that increased over the period (leading up to the housing bust)."
The authors contend that high loan-to-value ratio mortgages are a central hallmark of subprime lending. They define a high loan-to-value mortgage as one where the debt is greater than 90% of the value of the home itself. While I think it’s a stretch to label 90% LTV mortgages as subprime, the point that high leverage loans should not be considered “prime” is well taken. If we accept this definition, we immediately see the problem: CMHC and other insurers actively insures mortgages where the loan-to-value is 95% (i.e. a 5% downpayment). And even this is stretching it as we have seen that all big Canadian banks provide a loophole to allow the purchase of homes with 100% LTV as they offer 5-7% cash back upon closing….up until the end of October when new OSFI regulations kyboshed that practice.
The other popular definition of a subprime is based on credit scores. The Tal piece provided some interesting insights regarding the degradation of credit scores during the US bubble years.
The Federal Reserve Bank of Boston, in a report titled "Subprime Facts: What (We Think) We Know about the Subprime Crisis and What We Don't", defined subprime lending as the extension of mortgage credit to individuals with FICO scores below 620.
Once again, if we apply this to the Canadian marketplace, we find that CMHC will insure buyers with a 600 FICO score if they make at least a 10% downpayment. When the downpayment drops below 10%, CMHC will still insure a mortgage for anyone with a FICO score of 610. Both measures are below the threshold that typically delineates subprime from prime borrowers.
The point here is not that Canada has subprime lending to the extent that the US had. That’s never been my perspective. But I do think that the difference in lending standards has likely been vastly overstated at both the “prime” and the “subprime” tails of the lending spectrum.
Just as I believe that the term “subprime” does not transfer easily between Canada and the US, the same is true for “prime” loans where the vast majority of US prime loans even during the bubble years required two full years tax return, employment confirmation via phone call, and full physical appraisal were far more common than they are in Canada. This documentation requirement is vastly superior to “prime” loans originated via CMHC, in which only a job letter, pay stub, and electronic appraisal is required.
Furthermore, until recently, CMHC would insure high ratio loans made to self-employed individuals based on stated income, provided the income level was deemed “reasonable”. These are considered “prime” loans in Canada, but would most certainly have been labeled “non-conforming” in the US.
2) Was the US bubble “at its core a subprime story”?
It’s worth revisiting one of Tal’s key quotes from the report: “At its core, the US meltdown is a non-conforming story”. While subprime lending is an easy scapegoat for those looking for simple cause-and-effect explanations for the crash, that viewpoint is far from universally accepted among economists.
I personally would take the position that excess and unsustainable credit growth which artificially buoyed the economy and labor market, coupled with overly optimistic price expectations leading to various forms of speculation was at the core of the US housing bubble, as with virtually every other housing bubble in history. In this view, subprime lending was a symptom of this trend, not a root cause.
As I’ve said before, the tendency of credit booms to masquerade as economic booms ought to cause policy makers some sleepless nights given the current rise in household debt in Canada and the now cyclical (and in many cases all-time) highs in the GDP and labor market contribution of virtually every housing-related sector. Fed Chariman Bernanke’s now infamous 2005 interview in which he dismissed the possibility of a housing bubble because of strong “fundamentals”, namely a growing economy and labor market serves as an excellent example of how easily we can be fooled by credit-driven pseudo growth.
If, in fact, the business cycle is better thought of as a credit cycle, as I believe, then this goes a long way in explaining how the economy can look so strong when in fact it is being artificially supported by excess credit growth. So while the US bubble may have been unique in the prevalence of subprime lending, like all credit bubbles, it burst when the flow of credit slowed and negatively impacted the underlying economy, leading to a vicious spiral.
The credit boom in Canada rivals the US boom by every possible metric.
And while we can argue that somehow we’ve managed to achieve this by extending credit by-and-large to very credit-worthy individuals unlike our American cousins (doubtful), we can’t gloss over how this credit boom has affected the broader economy…..or the implications when credit growth slows dramatically, as I believe it will.
I’ll have more to say about areas of disagreement in a later post. I’d better leave it there for now.