A closer look at the house price forecasts from the Canada Mortgage and Housing Corporation (CMHC) shows they are not as alarming as they first seem. The worry, however, is that the very public warning of huge price declines becomes self-fulfilling.
A closer look at the house price forecasts from the Canada Mortgage and Housing Corporation (CMHC) shows they are not as alarming as they first seem. The worry, however, is that the very public warning from the crown corporation becomes self-fulfilling.
Representatives of public institutions understandably tend to err on the side of caution when it comes to discussing potential house price declines. This week Evan Siddall, CEO of CMHC (the crown corporation that provides government-backed insurance for high loan-to-value mortgages) took a different approach. Speaking before the Standing Committee on Finance, Siddal warned that house prices could slump by up to 18% over the next 12 months. That compares to our own forecast that prices will fall by 5%. (See here.)
A closer look, however, shows the CMHC’s forecasts are not as alarming as they first seem. First, the 18% decline is the CMHC’s worst-case scenario. In its best-case scenario, it is forecasting a lower, albeit still significant, fall of 8%. Second, and more importantly, CMHC is forecasting the change in the average selling price. By contrast, we are forecasting the change in the Teranet house price index, which essentially shows the change in the price for any given home.
This is an important distinction, because changes in average selling prices and changes in like-for-like house prices can be very different in downturns. As we have already seen in the home sales data for April, during market downturns the proportion of higher-value homes that are sold often falls significantly. That in turn pulls down the average selling price. For instance, the peak-to-trough fall in the average selling price during the Global Financial crisis (GFC) was, coincidently, also 18%. But the Teranet and MLS house price indices, which account for the composition effect among other things, showed much smaller falls, of 6% and 7% respectively. Even in the US where the selling pressure was more broad-based during the GFC, the peak-to-trough fall in the average selling price was 26% whereas the decline in the Case-Shiller index was 18%.
By our estimates, the CMHC’s forecasts translate into falls in the Teranet index, or like-for-like house prices, of between near-zero and 10%. At 5%, the mid-point of that range is the same as our own forecast.
CMHC will hopefully clear this up when they release their new forecast report next week, but we are concerned that some damage may already have been done. As the history of boom-and-bust cycles shows, individuals’ expectations play a big role in how house prices develop. With the CMHC’s alarming forecasts covered by all the major news outlets this week, some Canadians have probably become far more concerned about prospects for the housing market.
Admittedly, we are becoming more concerned about the risks to the housing market ourselves. We noted this week that rents in the big cities already appear to be falling fast. (See here.) With rental yields already very low, that is a big risk to house prices. The good news is that this hopefully won’t create much immediate selling pressure. Rents have been rising strongly in the past few years, so most investors would still receive higher rents even with the recent decline. But it is likely to be a problem for investors taking delivery of new condos this year. If rents in Toronto and Vancouver fall by 5% to 10% as we expect, many of these investors will face negative cashflows and may decide to sell their properties.
Coronavirus effects now clear in the data
As the official data are released with a longer lag than elsewhere, we had to wait a while to see the effects of the coronavirus outbreak. They became clearer this week, with retail sales plunging by 10% m/m in March and Stats Can saying that sales probably fell by another 15% in April. (See here.) Meanwhile, inflation turned negative for the first time in a decade in April. (See here.)
The week ahead
Next week, we will see what all that means for GDP in March and the first quarter. Our estimates broadly agree with the earlier nowcast produced by Stats Can, which pointed to a drop of 9.0% m/m.
GDP (%q/q annualised)
A huge fall, but a bigger one to come
A 9% m/m slump in March appears to have dragged GDP down by 10.5% annualised in the first quarter. We expect an even larger decline of 45% annualised in the second quarter.
The official data for March have been understandably poor. Retail sales volumes slumped by 8.2% m/m, manufacturing GDP appears to have fallen by 6% (see Chart 1) and wholesale trade GDP probably fell by 3%. There were probably also large falls in activity for many of the sectors for which we do not have specific data for. Indeed, Stats Can released a nowcast in April suggesting that GDP fell by 9% m/m. Our own crude estimates for those other sectors based on hours worked point to a similar sized decline, so that is what we are pencilling, although there is clearly a lot of uncertainty around the exact number. A 9% m/m drop would result in a 10.5% annualised contraction in GDP in the first quarter.
As the restrictions on activity only began in the middle of the month, and as they were tightened toward the end of March, we suspect that GDP fell by a larger 15% in April. Even assuming a 5% rise in May and a 10% rise in June, GDP would fall by 45% annualised over the second quarter.
Chart 1: Manufacturing Activity (% m/m)
Sources: Refinitiv, Capital Economics
Economic Diary & Forecasts
Time EST (BST-5)
Speech by BoC Governor Stephen Poloz
BoC’s Poloz & Wilkins speak before Senate Committee
No Significant Data Released
Current Account Balance (Q1)
GDP by Expenditure (Q1)
GDP by Industry (Mar)
Industrial Producer Prices (Apr)
Raw Materials Prices (Apr)
Selected future data releases and events
Manufacturing PMI (May)
Labour Productivity (Q1)
Bank of Canada Policy Announcement
International Merchandise Trade (Apr)
Change in Employment (May)
Unemployment Rate (May)
Ivey Purchasing Managers Index (May)
*m/m(y/y) unless otherwise stated
Sources: Bloomberg, Capital Economics
Main Economic & Market Forecasts
%q/q ann. (%y/y) unless stated
Unemployment Rate (%)
Overnight Rate, End Peri’d (%)
10 Yr GoC., End Period (%)
USD/CAD, End Period
Sources: Refinitiv, Capital Economics
Stephen Brown, Senior Canada Economist, firstname.lastname@example.org