Blogs & Comment

Four horsemen of a housing apocalypse

In the New York Times column, popular economist Robert Shiller suggests a handful of indicators that can presage a housing slump. Those same indicators suggest a benign situation in Canada.

People walk past new homes that are for sale in Oakville, Ont., on April 14, 2009. (Photo: Nathan Denette/CP)

What are the harbingers of a housing collapse? That has been a popular question for at least the last five years, ever since the Americans began experiencing sharp declines in the resale value of their homes. Those fretting about excessive leverage in this country (and I’ve been among them) have been confounded by the continued absence of a meaningful correction in this country; Indeed, the Canadian Real Estate Association reports that in April resale prices averaged about $373,000, up about 8% from a year earlier. They’re at or near all-time highs in most important markets across the country, while the U.S. continues to suffer. How might we know whether a reversal was in the offing here?

Recent comments by popular economist Robert Shiller of Yale University, although intended for an American audience, provide some clues about what data Canadians might track. America’s housing crash is typically dated from 2007, when prices began their precipitous slide. But in a recent column for the New York Times, Shiller argued the market actually peaked two years earlier. If he’s right, the chief implication is that if you’re simply watching prices, you may miss the true signs that disaster is at hand.

Shiller regards the bubble (1997-2006) and subsequent bust (2007-present) as not simply the product of mistakes by the Federal Reserve or malfeasance by Wall Street elites, but rather as what he calls a “social epidemic.” His basic argument is that rising prices over a long period of time elevate expectations, which feed a “contagion of optimism” resulting in a speculative bubble that must ultimately correct.

Shiller’s first indicator of disaster is hardly novel. As many others have done, he first points to the meteoric rise in U.S. home prices during the boom years. Between 1997 and 2006 the Case-Shiller National Home Price Index (which Shiller himself helped develop) rose 131%, which translates to nearly a 10% average annual gain. Significantly, interest rates on mortgages were typically far lower than that, which meant the vast majority of homeowners racked up unrecognized gains on real estate with no effort whatsoever. The run-up in prices was important primarily because it shaped expectations: This period was “long enough for many people to become accustomed to the pace and to view it as normal,” Shiller writes.

He and colleague Karl Case are in a position to know, because they survey Americans every spring to assess attitudes toward home-buying. Among other things, their survey asks how much respondents expect property values to increase on average each year over the coming decade; Their Spring 2005 survey found that respondents expected 7% annual gains through 2015. Had that come to pass, prices would have doubled again during that period. “Something was very wrong with this picture,” Shiller writes, “but few noticed it.”

How far have Canadian prices elevated? The closest thing Canada has to the Case-Shiller index is the Teranet—National Bank House Price Index. Its data goes back only to 1999. Between February 1999 and the most recent data in March, that index rose 109%, or an annual compound average of around 6.4%. Although dramatic and extended, the price run-up here has been less pronounced than that witnessed south of the border.

What impact has this run-up had on expectations? I can’t find any Canadian survey that asks a question as specific as that posed by Case-Shiller. The closest I could find comes from the Canadian Association of Accredited Mortgage Professionals. Since 2005 it has conducted biannual surveys of consumers, with an eye to understanding what’s happening in the residential mortgage market. Its most recent report, published last month, stated:

Expectations about house prices were weak during the recession of 2008/2009, but have become considerably more optimistic. When asked “to what extent do you think housing prices in your community will go up or down in the next year?” just 8% expressed negative opinions this time (giving scores of 1 to 4 out of 10). Almost one-half (46%) expect prices to rise to varying degrees (ratings of 7 to 10) and 46% gave neutral answers (5 or 6 out of 10). The average rating of 6.36 is well above the midpoint of 5.5.”

That doesn’t tell us what Canadians consider to be a “neutral” or “strong” increase. But CAAMP also asks its respondents whether they think it’s a good time to buy houses in their community. In this year’s survey 44% thought it was (22% who thought it wasn’t and 34% offered neutral responses). Notably, these responses were also slightly less positive than those received in the past two years. Although we can’t be certain, these survey responses suggest most Canadians aren’t expecting 7% annual increases on their homes. Notably, however, a great many would be taken by surprise if a reversal happened now.

Shiller points to two more indicators. They both come from the National Association of Home Builders, and are available only by subscription. One tracks the volume of prospective buyers traipsing through new houses, a key indicator of people’s willingness to buy. The other is housing permits, which are authorizations from local government agencies to construct a specified house on a given site. In a podcast discussing his recent column, Shiller said that while prices continued rising after 2005, that year these two key indicators began falling rapidly. “It isn’t the crisis that caused this,” Shiller told podcast host Jeff Sommer, assistant business editor at the Times. “This is what caused the crisis. The first signs of the subprime crisis occurred two years after the peak.”

Again, CAAMP’s statistics shed some light on the current Canadian situation. Typically between 4.5% and 5.5% of Canadians buy homes in a given year, the organization claims. The latest survey this spring found buying intentions at the low end (4.5%) of this range, but still considerably higher than the depressed levels (3.2-3.6%) seen during the recession. There’s been no abrupt drop-off in interest.

What about housing permits? According to the Canada Mortgage and Housing Corp., last year there were nearly 43,500 residential permits issued, higher than during the previous two years. We’ve also not seen a sharp retreat in housing starts. So if there’s reason to panic, builders haven’t figured it out yet.

Taking Shiller’s suggestions at face value (and perhaps a few liberties as well) there are seemingly few reasons to anticipate a crash in Canadian residential real estate during the next two years. But isolated from all else, his “social epidemic” theory seems less than satisfying as a comprehensive explanation of America’s housing debacle. For my part, I’ve long believed consumer leverage ratios are among the best ways to gauge how vulnerable housing markets are to abrupt declines. Still, having been among the prescient voices raising concerns about America’s housing market prior to 2007 (and also the stock market prior to its collapse), Shiller’s comments about bubbles of any kind continue to find an attentive audience.