We recently had an illuminating chat with Genworth Financial Canada Chairman & CEO Brian Hurley. Topics of conversation included first-time homebuyers, housing risk and causes of default.

We began with first-time buyers, which account for a large slice of the market. In fact, they comprise a veritable army of buyers, one that’s vital to the health of residential real estate.

Hurley went on to share some noteworthy points about mortgage defaults. For example, he noted that default frequency tends to correlate most closely with unemployment. However, the number one factor that affects an insurer’s default losses is when home prices drop more than expected. That doesn’t necessarily cause the frequency of claims to skyrocket, but it does cause the size of claims to balloon, and that can present a problem.

Hurley explains the above and more in the comments that follow…

On the size of the first-time buyer market…

Brian: I think a good proxy is looking at those individuals who are getting a high loan-to-value (LTV) loan. A big portion of that would be first-time homebuyers and that’s what we see in our insurance volumes year in, year out. Just to be more specific, typically, and this is over a long period of time, the high loan-to-value component of the marketplace was north of 40%. So it was a big component of the market being high loan-to-value loans. This year we are seeing a bit of a changing dynamic for the first time in a while, in the first quarter anyway. It’s down a bit to about a third of the market being high loan-to-value lending, but it’s always been a big component and within that component I’d say that the profile would be predominantly first-time homebuyers.

On the average down payment for a first-time buyer…

Brian: Going back to our own database and our own portfolio, which I think is a pretty good indicator, our average loan-to-value is about 90%, so the average down payment for this group would be about 10%. A lot of variability there, but I think that’s a pretty good benchmark.

On the number of people putting down only 5%…

Brian: It’s been shrinking, especially since some of the government regulations came out over the last couple of years. Going back to a year like 2007, for example, it was a very significant piece (of the market)…Right now I’d say with a big chunk of our business being 10% down, LTVs above 90% have dropped dramatically over the last couple of years. And even when you look at those individuals, they exhibit very good buyer profiles; strong credit scores, they have a good earnings base, and their debt service levels—which are a concern for the market as you well know—are usually quite reasonable. So, even when you do see someone with 5% down, it’s a much stronger borrower profile.

On what a “reasonable” total debt service ratio looks like…

Brian: It’s going to sound like a weasely answer, not that I’m trying to be, but a lot of it depends on what your outlook is. So, for us for example, we have a program that looks at folks who are newly graduated from medical school who may not have a big earnings base now, but have a pretty good prospect of significant earnings growth for the next five, 10 years. So, we would put them on the upper limit of our debt service levels versus someone with a dual income who may be barely making it, if you will. We keep them at the lower end to give them some more room for rate shock and for life occurrences. It really does depend on the profile and where you are coming from and what your earnings potential looks like going forward.

On the average purchase price of first-time buyers…

Brian: Again it varies market by market, but on a national basis CREA, which is the Canadian Real Estate Association, is about $365,000 average for a home. For us, for the first-time homebuyer piece, that would be about $295,000 on a national basis…our first-time homebuyer piece (in Vancouver) would be $453,000…a big difference from the national average…

On how a fall in prices of high-end homes could impact homes bought by first-time buyers…

Brian: It will have a trickle-down effect, no doubt about it. How much? I guess that’s up for debate. Our view is, if market prices are corrected 20% let’s say over a multi-year period, the trickle-down effect on our segment of the market would likely only be a 3-5% home price depreciation…Affordable [homes] would not feel the full impact of a correction on the high, high end.

On the number of high-end buyers who pay cash…

Brian: If you look over the last year at the high, high end, (like in Vancouver) I would say a big chunk of that—north of 50%—was cash transactions…When you look at 2011, especially the first half of 2011, there was not a lot of sales activity, but a lot of high, high end activity and a lot of cash transactions, which gave the overall Vancouver market, I would say, almost an artificial bump up, because they influence the market so much.

On the chances of first-time buyers going “underwater” and being prevented from selling if home prices fall…

Brian: It’s something that we’ve got to watch…The watch markets are of course the GTA in their condos and Vancouver on the high end. We still have our eyes on those same two areas…The view today is, you don’t have that built-in home price appreciation that we have enjoyed for many, many years up until 2007, where all of us were getting, you know, 5%, 6%, 8% almost built-in annual home price appreciation. The behaviour of those folks since 2007, when there’s been negligible appreciation, or even home price declines in some markets, the behaviour is still very strong. People pay their debts when they can, people will stay in their home longer versus upgrading, which is fine. We don’t see them stretching on the debt service levels like Ottawa is warning us about. We don’t see that coming through in our portfolio; as a matter of fact we see some good self-discipline there.  So I think the behaviour is you’ll stay in your home longer, you won’t be getting an upgrade and, if you’re forming a family or something, it may be crowded for a few more years, but we don’t see delinquencies rising.

On how Canadian homeowners would react if they owed more than their house is worth…

Brian: (If you’re) talking about the States where people can throw in the keys and move on, that’s a very different dynamic than what we have here of course. But our correlation is, unemployment drives delinquencies for our business. If people are employed, if they kept their job, they’re going to make their mortgage payments regardless of what’s happening with house prices. The view is that they’re going to hang on and weather the storm, and for the longer term this will be a smart investment for them, even though it may be stagnant or down in the short term. Home price decline in our insurance business increases our claim size. So when there is a default, likely a default because of unemployment, the size of the claim that we pay will be larger because of home price declines, we can’t recover as much, or we don’t have that built-in home price appreciation.

So again, delinquencies correlate to unemployment. As we’re seeing some slight move in unemployment, we’re seeing some slight positive movements on delinquencies, which is good. It’s important to note that from a claims perspective, as we talked about, it’s the home prices that are going to be the biggest determinant of our claim size. So, if we see variability around home prices, that’s where we’re going to get hurt, probably more so than delinquencies.

On how underwriting deals with the chance of higher rates…

Brian: When we underwrite a borrower, if they’re going to be a variable rate for example, we’re going to tack on an extra 300 basis points of ratio with the full anticipation that rates will go up. We know that (rates will rise) for a fact here in the marketplace. So, you need to treat that borrower with some understanding that, in at least their first five-year term, rates are going to increase…How much you underwrite for…could be anywhere between 200 and 300 basis points…Remember, the worst thing for us in the industry is defaults. We want to get people into homes and keep them in their homes. That’s important. And in a low-rate environment, almost an artificially low-rate environment, we need to make sure that these individuals are prepared for that increase whenever it comes.

On using a higher qualification rate than required by regulators…

Brian: We saw with BMO a few months back, and a lot of these very, very attractive rates that are out there. I think it would be doing the borrower a disservice if we let them qualify for 2.99%, knowing that over the next five years or sooner they’re going to see some (rate) pressure…when they go to renew or upgrade, or whatever the situation might be. So I think it’s prudent in a low-rate environment to look ahead and anticipate some rate increases.

For us, if we see someone stretched too much in a debt service level, or whatever it might be…we work with the lender and say, you know what, these individuals are stretching themselves too thin, or these individuals can’t withstand an increasing rate environment. We suggest that we qualify them at 200 basis points more, or whatever it might be, or we suggest they take a lower LTV or even get a smaller home or a smaller loan. So, we make sure that the (borrowers) that we‘re insuring for our lender partners are going to make it.

On whether 2.75% is enough of a default insurance premium for someone with only 5% down…

Brian: You know, good question. It’s important to look at this business and mortgage market over cycles. You can’t have a snapshot of 2012 or 2011, you need to look over a 10-year period and when you look over that 10-year period, for us, there were many years that the market was going very, very well. Home price appreciation was very strong, very little default, and therefore very little losses for the insurers. We enjoyed those years for quite a while. Over the last few years with delinquencies being up, and more unemployment pressure, we’re paying claims.  And, in a way, that’s just what the model is supposed to do. So yeah, (2.75%) is adequate. On some years, like last year or the last couple years, we’re not as profitable.  We might not have those margins that we once enjoyed, but when we look back, the pricing model works. This is enough to build adequate reserves, and this is also enough to generate sufficient returns for the private sector so that they will invest in our business.

Those prices haven’t changed much, for a while. And I’ll say those prices, and the dynamics around the underwriting and the insurance, worked well during this economic downturn that we’ve been experiencing.

On the extent to which Genworth has modelled the risk of a housing correction…

Brian: We’ve modelled it. It’s a hot topic for us and for the industry, of course. With our regulator we’ve modelled this every which way from Sunday—to national decline of 25% over a four-year period of time, or unemployment reaching 15% for an extended period of time, and every iteration in between. It’s important; I think it’s been a healthy exercise for the overall industry and for us.

Rob McLister, CMT