« Liquidity, solvency, fire-sale and fair prices, and TARP: a simple model | Main | The federal government and the CMHC enter the fray »


Feed You can follow this conversation by subscribing to the comment feed for this post.

It's good to see that data, but I don't interpret it quite as optimistically. The US delinquency rate is currently about ten times as big as Canada's. But then US house prices peaked about a year earlier than Canada's, and our house prices have not fallen as far yet. So Canada might just be catching up to the US. Also, even though Canadian homeowners have higher equity at current prices, that will partly just reflect the fact that US house prices have fallen further than Canadian, so far. Even so, it would take a further 40% reduction in Canadian house prices to bring our 70% equity down to the 50% equity of the average US, and US house prices haven't fallen (on average) 40% more than Canadian house prices. So it does look as though Canadians are less mortgaged, even allowing for the fact that our bubble peaked later than theirs. Which is some good news on a dark day.

What are the underlying structural differences between Canadian and US housing/mortgage markets? I can think of:
1. They can deduct mortgage interest from taxable income; we can't (so we have a bigger incentive to pay off our mortgages).
2. Most of their mortgages are non-recourse; ours aren't (so we have less incentive to walk away from negative equity homes).
3. We are not allowed 80% or higher mortgages without costly mortgage insurance (so fewer subprime mortgages).
4. Our banks have faced less political/legal pressure to give mortgages to less riskier borrowers.
5. Does it make a difference that all US mortgages are effectively open mortgages (they are allowed to refinance without penalty)?
6. Anything else?

It'd be interesting to do the counter-factual of 'what if Joe Clark had won in 1980?' You may recall that one of the measures in the defeated PC budget was to make mortgage interest deductible. I remember talking to an American a few years ago who remarked that paying off his mortgage was - under the circumstances - one of the more foolish financial decisions he could make. Since Canadian income taxes are higher, there would be a correspondingly higher incentive to take on big mortgages.

I don't think mortgage interest deductibility is a factor really. As long as a mortgage is not in arrears the lender generally comes out ahead if the mortgage is paid off within the term. To be quite honest I cannot think of a normal situation where you are farther ahead income tax wise by not paying off the thing and making interest payments. Even with what amounts to a government subsidy, it still costs you in the long run, more (unless the entire interest is subsidized).

I think point #3 and #4 mentioned above are more significant and probably likely to bear out to be explanatory factors amongst a whole littany of government and business shady relationships.

langman, I think mortgage interest deductibility does reduce the incentive to pay off the mortgage:

Suppose my only asset and liability is a house and mortgage, suppose the interest rate on the mortgage is 6% and my marginal tax rate is 50%. With mortgage interest deductibility my after tax rate of interest becomes 3% rather than 6% without deductibility, so I have less incentive to save, so pay down my mortgage more slowly.

Now suppose I can also save by buying bonds which also pay 6% (taxable). With mortgage interest deductibility I am indifferent between buying bonds and paying down the mortgage (both give me an after tax return of 3%). With no deductibility I earn 6% after tax from paying down the mortgage, and 3% after tax from buying bonds, so pay down my mortgage. Raise the interest rate on bonds to 7%, and I would buy bonds rather than paying down the mortgage with deductibility, but would still pay down my mortgage without deductibility.

Now suppose I can earn 6% tax free by saving inside an RSP (assume my marginal tax rate will still be 50% when I withdraw the funds). Under deductibility, I save inside the RSP rather than paying down the mortgage. Without deductibility, I am indifferent between the two.

Which countries around the world have deductibility? Do those countries have worse house bubbles, bigger mortgages, and more mortgage defaults?

I agree. Mortgage interest deductibility must be a factor contributing to higher mortgage balances. I assume the sub-prime crisis was/is largely due to defaults among the less wealthy. Sure, some middle and upper income folks have been caught up as well, but I think it must be the less wealthy that make up a large percentage of the defaulting homeowners. So how much did interest deductibility affect them? In Canada, few homeowners are in a 50% tax bracket. The MTR for most is far less. While I agree that intrest deductibility will offer an incentive to carry a larger balance, I don't agree that Canada's situation is greatly improved by the absence of this incentive. If all other factors are removed how much would mortgage interest deductibility contribute to the mortgage default rate? I like Nick's list which appears to be randomly sorted. Should this list be sorted by each factor's influence, I think the deductibility item would be very low on the list. Some other ideas:

7. Risk aversion
8. Population density

Does anyone know where the "Canadians have on average 70% equity in their house" data come from?


They can deduct mortgage interest from taxable income

Only about 30% of eligible U.S. tax filers actually take the deduction as interest deductibility is not cumulative to the standard deduction. It is estimated that the deduction costs the U.S. Treasury about $90 billion per year. Basically, it is a sop to trophy homeowners. The deduction goes to Manhattan, New York, not Manhattan, Kansas; Lafayette, California, not Lafayette, Louisiana.

Enlightened Canadian homeowners do take advantage of mortgage interest deductibility for investment purposes.

@ Rowe,

I agree, but it's still in your interest to pay the mortgage at the terms or you would lose the capital. What I was saying was that this doesn't hurt the lender, especially if as interest rates are rising they can raise their rates in association with the rise in bonds. Therefore I don't think this is a large factor in the number of arrears in that market, if it actually contributed to arrears then I would agree it hurts the lender.

Does anyone know where the "Canadians have on average 70% equity in their house" data come from?

As a crude approximation there are 8.5 million homeowners times $300,000 average price minus $850 million in mortgage debt.

As a Canuck who lived in the US for 11 years, 94 - 05 I'll provide some info that expands on what Nick Rowe provided.

1. They can deduct mortgage interest from taxable income; we can't (so we have a bigger incentive to pay off our mortgages).

In the US you have a choice as far as deductions on personal income tax is concerned.
You can take a standard deduction or you can what is called, itemize your deductions. Itemizing (listing) allows the mortgage interest, medical expenses etc to be deducted from your income. You would only itemize if the sum itemized was greater than the standard deduction. So if the mortgage interest was less than the standard deduction you wouldn't choose to itemize. You do not get to take a personal deduction as well as deduct mortgage interest. There are caps on how much you can deduct so at times you will not receive the entire deduction. Medical expenses don't deduct from income until the amount exceeds a certain percent of income. For us it would have required approx. 13K in expenses before any deduction occurred.

3. We are not allowed 80% or higher mortgages without costly mortgage insurance (so fewer subprime mortgages).

Mortgage insurance was required at 90% or higher in the US. This may have changed in recent years but back in 90's that was the case.

5. Does it make a difference that all US mortgages are effectively open mortgages (they are allowed to refinance without penalty)?

There is no penalty to refinance but the costs associated with refinancing can be substantial.

There is an origination fee that is typically 1% of the mortgage amount that is paid to the mortgage/banking institution. It can be higher than 1% but not usually. 200K mortgage = $2,000 origination fee. This does not include prepaid taxes, legal fees etc. 3K to 6K in closing costs for a new home is the norm. I believe our closing costs on a 200K home in 95 was around $3400. If a buyer could only provide the minimum down payment, but didn't have money to cover the closing costs, the closing costs were added to the cost of the house and the builder would pay the closing costs.

Just my experience living in the US.

Is this guy saying that what Harper said during the debates is... true? No way. No $%^&* way.

Harper was saying that 10 months ago. At that time, all the opposition wanted to talk about was Schreiber & Mulroney. Now they are dumping on him because he is not panicking like them. And the media is helping them by blowing this crisis out of proportion. Sad.

I'll vote for no one except Harper.

Now suppose I can earn 6% tax free by saving inside an RSP (assume my marginal tax rate will still be 50% when I withdraw the funds). Under deductibility, I save inside the RSP rather than paying down the mortgage. Without deductibility, I am indifferent between the two.

Nick, you are now out of your league. The only way a tax deferred account is tax effective is if the marginal rate of taxation (MTR) is less on withdrawal than it is on deposit. But we won't complicate the matter with MTR on dividends or capital gains.

Now I can't find you at Carleton or google. Why not?

Nick Rowe is a member in good standing of the academic community. For my part, I'm pleased (ecstatic, actually) to have someone like you post here, because you clearly know your stuff, and I for one have revised my thinking based on your comments.

Nick and I are both willing to defer to analyses based on information better than that available to us.

Thank you, Stephen. Nick received many comments at Thoma's Economist View on this post. Unfortunately, he did not respond to a wider audience.

However, when the canal freezes, he can skate down to uOttawa and cafe latte with one of my kids at the National Arts Centre.

I'm a LLB/MBA, so forgive me for my indiscretions on your economics blog.

Nick posted here, and he responded to comments posted here. All people had to do was follow Mark's link to the original post.

Hi marmico: yes, I really exist. Try "Nicholas Rowe" on Google (or whatever).

I disagree on the RSP. Suppose I have a 50% MTR for all periods, and invest for 10 years at a 6% interest rate.

Inside an RSP my after tax return after 10 years is:


If I invest outside an RSP, and remember I have to withdraw half of my annual interest earnings to pay tax, so my investment is compounding at 3%, and my after tax return after 10 years is:


Whether you pay the tax up front or on withdrawal doesn't matter (as long as the MRT is the same). The benefit of an RSP is that money inside an RSP accumulates at the before tax rate of interest, rather than at the after tax rate of interest. It's all in the compounding.

To put it another way, the new tax free savings account recently announced (I've forgotten the exact name), where you don't pay annual income tax on the interest earned, is indeed equivalent to an RSP, provided your MRT stays the same. But both are better than regular savings, where you pay tax both when you earn the original $100, and each year on the annual interest income.

As always, I am not 100% sure I am right. I used to think the same as you about RSPs, then 20 years ago a colleague sat me down and explained I was wrong, but I didn't understand him until I sat down and slowly, eventually, figured it out for myself, with numerical examples and then the formula above.

I wondered whether to reply to the comments on Mark Thoma's blog. But there were so many, I got scared of trying. Plus the comments seemed to be arguing all points nicely among themselves, so I just let it run.

6. MBS as a percentage of outstanding mortgage debt was never higher than 20% in Canada. It hit 50% in the US.

7. Canada's housing body the CMHC was never as aggressive as their US cousins (Fannie, Freddie, and Ginnie) in introducing new mortgage products. For instance, the CMHC introduced 100% mortgage products in 2006 (no down payment). Fannie and Freddie had pioneered those products in 2001 and 2000 respectively.

The CMHC started to do 5% down mortgages in 1999, Fannie and Freddie had been offering these since 1993 and had already moved up to 3% down mortgages by 1996.

What should we make of this? Seems a questionable move to me, at the very least.

"I'm a LLB/MBA, so forgive me for my indiscretions on your economics blog." -marmico

Does that explain the apparent preference for 'authority arguments'? (Sorry, couldn't resist.)


On a more substantive note, I would simply add that the real economic impacts of the wind down of the western Canadian, resource-based economy have yet to fully manifest themselves. Western Canadian real estate markets will follow.

Cutting the federal valued-added consumption tax (GST) during this last boom was unfortunate -- bloody cynical if you ask me -- and will look even more unfortunate in the rear-view mirror.

Josh: I am trying to get a better understanding of that plan announced today (described in the Globe and Mail article you link).

As far as I can see, the government (or the CMHC?) buys some of the mortgages held by the banks, giving the banks T-bills in exchange. Presumably the bank then sells those T-bills in the open market, and uses the proceeds to finance new mortgages etc.

The major problem in the financial markets is that interest rates on safe short loans (like T-bills) are very low, while interest rates on long or riskier loans (like mortgages) are very high. So the government is increasing the supply of T-bills and increasing the demand for mortgages. It's a modern variant of "Operation Twist", which was a 1960s(?) US Fed plan to twist the term structure of interest rates by buying long bonds and selling short bonds.

I think it makes sense, as a temporary measure. But as a long-term plan, it sounds a bit Fanny and Freddie?

I just started another post on this.

It is unlikely that the rate of default in Canada will ever get as high as it did in America, given that lending standards are tighter. However, even though the rate of default did not get above 1% in the 1990s, it did not stop the price of homes from falling. In Toronto, home prices fell from ~$300k to nearly $200k later in the decade. On the west side of Vancouver, prices fell from $770k to $550. The lack of a subprime mortgage market did not stop over-valued house prices from falling.

The US is 10x bigger than us, so too will be their foreclosures, people underwater with mortgages when our conditions are comparable.

BUT, they are not - they are in the middle of their housing collapse, we are at the beginning.

the middle of our own housing bust will be the only time that we can make comparisons of our economy, banking system, etc to theirs. TO say that 'we are not the US' is preposterously premature, something you'd expect Harper to say, not people who actually think about what is going on

Nice to read your blog as Alberta seems to be singular in its lack of blogs and commentators offering the other side of the housing story. I just got back from a little introductory home buyers session for UofA alumni courtesy of the CMHC, two local realtors, and a local mortgage broker who works with the realtor. All, aside from the CMHC guy, painted a fairly breezy picture of real estate in Alberta (CMHC guy was neutral).

I asked a few questions about what happens to a new homebuyer that needs to get out of their mortgage for a career issue or whatever, if they have only 5 per cent down and the market has dropped 15 per cent. The answer from the mortgage broker: not very likely to happen, but you could rent it out. I wasn't so confident - if you're paying an inflated mortgage, how will you cover your payments with a lower rent?

The non-CMHC folks also told the crowd that Canada wasn't exposed to subprime mortgages. That didn't seem to ring 100% true with me for some reason, although though this has generally been accepted in the local media.

Tonight I looked at an RBC economic update from the spring (http://www.rbc.com/economics/market/hi_city.html) that stated "while U.S. sub-prime mortgages account for roughly 14% of the market, Canada’s sub-prime mortgage market is negligible, with the highest estimates showing a moderate 5% share." One third of the U.S. problem in Canada still seems like a real a problem to me.

I'm going with my gut for now, and holding on to the down payment until I feel like we've seen bottom. The UBC Sauder graphs shows a wildly steep climb of the last couple years compared to historical boom-bust patterns from the peaks of ~1980 and ~1990. Another graph with nominal prices in Edmonton can be seen here: http://spreadsheets.google.com/ccc?key=p7pJfatJkrj79jChu53FRUw&hl=en (compiled by www.chrisdavies.ca).

Am I crazy and/or paranoid?

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad