Will a housing downturn crash the economy?
My subscription
...
Filters
My Subscription All Publications

Will a housing downturn crash the economy?

  • We expect higher interest rates to cause a 10% fall in house prices over the next 12 months and an even steeper fall in residential investment. With the rest of the economy still recovering from the pandemic and benefitting from higher commodity prices, that weakness should cause GDP growth to slow sharply rather than turn negative, but the downside risks of recession are rising.
  • The Bank of Canada’s hawkish shift has caused mortgage rates to surge, with five-year fixed rates doubling since September, to above their pre-pandemic peak from 2018, and variable rates heading in that direction if the Bank follows through with its plans to raise the policy rate to at least 2.5%.
  • When interest rates rise, home sales are usually the first domino to fall and this time is no different. There were large declines in home sales across Canada’s major cities in April, with those in Toronto and Vancouver plunging well below pre-pandemic levels. Anecdotal reports suggest that sales fell even further in those smaller towns and cities that experienced the largest house price gains during the pandemic.
  • With sales plunging, it is no longer a question of whether house prices will fall, but rather how much will they fall by? During previous tightening cycles, the popularity of fixed rate mortgages, and long fixed payment periods on variable rate mortgages, helped to prevent forced selling and large price declines. Nevertheless, because house prices have risen so far beyond a level that can be justified by underlying fundamentals, we suspect that, even in the absence of forced selling, prices will fall by 10% during this tightening cycle.
  • So long as house prices do not fall by much more than that, the largest impact of the housing downturn on the economy will be through residential investment. We are sceptical of the view that an apparent housing shortage will keep residential investment strong even as borrowing costs continue to rise. The reality is that, when home sales fall, new construction investment normally follows suit, for the simple reason that most developers fund their projects using pre-construction sales. We expect a downturn in residential investment to cause GDP growth to slow below its long-run potential in 2023, which will ultimately prevent the Bank from tightening policy by as much as markets are pricing in.
  • Our forecast that GDP growth will slow, rather than turn negative, is based on the view that the rest of the economy will continue to do well as it recovers from the lingering effects of the pandemic and benefits from higher commodity prices. But if house prices fell by much more than we expect – which clearly should not be ruled out given their elevated level – a recession would be almost inevitable. In that scenario, there would be a much broader tightening of financial conditions, with equities weakening, banks curtailing lending and credit spreads widening. Those tighter financial conditions would essentially sink all boats, weighing on activity elsewhere in the economy despite otherwise positive fundamentals.

Will a housing downturn crash the economy?

The negative impact of higher interest rates on house prices and residential investment leads us to expect GDP growth to slow below its long-run potential in 2023 and, as we explain in this Focus, even our below-consensus forecasts will prove too optimistic if house prices fall by more than we anticipate.

Bank rapidly tightening policy

The Bank has made no secret of its plan to continue rapidly tightening policy. Governor Tiff Macklem has all but confirmed that the Bank will hike its policy rate by 50 bp again in June, to 1.5%, and says that the Bank wants to get the policy rate to neutral territory, which policymakers judge to be between 2% and 3%, as soon as possible. Market participants believe the Bank will need to go even further, with overnight index swaps (OIS) now implying that the policy rate will rise to 3.25% in 2023. (See Chart 1.)

Chart 1: Policy Rate (%)

Sources: Refinitiv, Capital Economics

The Bank’s hawkish shift – markets were pricing in a terminal policy rate of 1.75% at the start of the year – has caused some extreme moves in mortgage rates. The so-called “deep discount” five-year fixed rate, the rate available to the most creditworthy borrowers, is now above its pre-pandemic peak, at 4.0%, and variable mortgage rates will quickly head that way as the Bank hikes further. (See Chart 2.)

Chart 2: Interest Rates (%)

Sources: Refinitiv, Capital Economics

This raises some serious questions about housing and the economy more broadly. Can house prices withstand a return to pre-pandemic mortgage rates, even though they have surged by more than 50% in the past two years? (See Chart 3.) And to what extent does the elevated level of residential investment, the most interest rate sensitive part of the economy, put the broader economy at risk? (See Chart 4.)

Chart 3: Average House Price ($, 000)

Sources: Refinitiv, Capital Economics

Chart 4: Components of Residential Investment
(% of GDP)

Sources: Refinitiv, Capital Economics

Plummeting sales to pull down house prices

Home sales are usually the first domino to fall when interest rates rise, and this time is no different. The local real estate board data for April showed large falls of around 30% m/m (in seasonally adjusted terms) in Toronto and Vancouver, and anecdotal reports suggest that sales fell by even more in the smaller cities in Ontario and British Columbia that saw the largest price rises during the pandemic.

As new listings held up, the sales-to-new listing ratios for those cities are now pointing to much weaker house price inflation. In Toronto, the ratio already points to modest house price declines (see Chart 5), and Vancouver seems to be heading in that direction.

Chart 5: Toronto Sales-to-New Listing Ratio & House Prices

Sources: Refinitiv, Bloomberg, Capital Economics

With variable mortgage rates set to rise further as the Bank hikes its policy rate, home sales are likely to continue to slide. Our big concern is that, because sales surged so sharply during the pandemic, they will now fall by even more than the rise in mortgages rates alone might seem to suggest. (See Chart 6.)

Chart 6: Mortgage Rates & Home Sales

Sources: Refinitiv, Bloomberg, Capital Economics

Our assumption is that home sales will ultimately fall by 50% from where they were in the first quarter, to 30,000 per month, which would be 25% lower than their pre-pandemic average from 2019. That will translate into an immediate proportionate fall in the ownership transfer costs component of real residential investment, and nominal ownership transfer costs will weaken by even more if house prices fall.

Historically, Canada has avoided substantial house price declines. Even during the early 1980s Volcker Shock, when five-year fixed mortgage rates surged from 10% to 23%, the economy contracted for five consecutive quarters, and the unemployment rate almost doubled from 7% to 13%, the peak-to-trough decline in house prices was just 9% – although the fall was much larger after adjusting for the change in consumer prices. (See Chart 7.)

Chart 7: Change in House Prices Following Tightening Cycle (%)

Sources: Refinitiv, Bloomberg, Capital Economics

The main reason why house prices have been resilient in previous tightening cycles is the absence of widespread forced selling. In the 1980s, that was due to the popularity of fixed-rate mortgages, but even today the risk of forced selling is reduced by the fact that many variable rate mortgages have fixed payments for a pre-determined period of time, normally five years, with any rise in interest obligations offset by reduced principal repayments.

Nevertheless, given that house prices entered this tightening cycle at a much higher level relative to fundamentals than ever before, we still judge that house prices will need to fall by around 10% to balance supply and demand.

New construction to fall alongside sales

Many commentators have cited an apparent housing shortage as reason why residential investment will remain strong even as borrowing costs rise, but the reality is that, when sales fall, new construction investment normally follows suit, for the simple reason that most developers fund their projects using pre-construction sales. (See Chart 8.)

Chart 8: Home Sales & Housing Starts

Sources: Refinitiv, Capital Economics

There are a couple of reasons why the reduction in new construction investment, which includes new projects and continued progress with existing ones, should at least be more gradual this time. The first is that, compared to previous tightening cycles, a much larger share of construction is now concentrated in high-rise projects (see Chart 9), which take longer to complete. So even if housing starts fall quickly, investment in existing projects will remain relatively high until those projects are completed.

Chart 9: Units Under Construction (000s)

Sources: Refinitiv, Capital Economics

The second reason is that an increasing share of housing starts are now purpose-built rental buildings, which are not reliant on pre-construction sales for funding. (See Chart 10.) Nonetheless, tighter credit conditions are still likely to lead to a reduction in these projects, particularly as the elevated apartment vacancy rate casts doubt on the idea that there is a severe housing shortage. (See Chart 11.)

Chart 10: Urban Housing Starts (000s)

Sources: Refinitiv, Capital Economics

Chart 11: Rental Apartment Vacancy Rate (%)

Sources: Refinitiv, Capital Economics

Residential investment to fall below pre-covid level

The final part of the puzzle for residential investment is renovations. While many renovations are linked to home sales, the historical correlation between changes in renovations investment and sales, or even house prices, is weak (for those who are statistically minded, the r-squared values are close to just 0.1).

Rather than a weakening housing market, the biggest risk to renovations is probably that the jump in demand for home improvement projects during the pandemic will soon run its course, particularly amid higher borrowing costs. We therefore assume that renovations spending will return to its pre-pandemic level by mid-2023 and, to the extent that people brought forward demand during the pandemic, there is a risk it could fall even further.

Altogether, we forecast that real residential investment will briefly fall below its pre-pandemic level in early 2023, mainly reflecting the weakness of home sales, before staging a modest recovery from the second half of next year. (See Chart 12.)

Chart 12: Real Residential Investment ($bn)

Sources: Refinitiv, Capital Economics

Can the economy handle a housing downturn?

While the outlook for residential investment is poor, there are a few reasons why we expect the economy to continue to grow in the coming years, albeit at a slower pace than the consensus forecasts imply.

One reason for optimism is that, while residential investment surged during the pandemic, the same was not true for the other most interest rate sensitive parts of the economy, durable goods spending and machinery & equipment investment. (See Chart 13.)

That was largely due to the global supply shortages of semiconductors, which reduced the availability of motor vehicles. As these supply shortages ease, we expect spending on vehicles by consumers and firms to rebound even as borrowing costs rise. For households, there is scope for a rebound in motor vehicle consumption to offset the likely fall in spending on furniture and household appliances as the housing market weakens. (See Chart 14.)

Chart 13: Interest Rate Sensitive Components of GDP
(% of Nominal GDP)

Sources: Refinitiv, Capital Economics

Chart 14: Durable Goods Consumption ($bn)

Sources: Refinitiv, Capital Economics

As well as pent-up demand on the part of firms and consumers, the economy should also benefit from the surge in commodity prices earlier this year, which will support both machinery & equipment investment as well as engineering investment, which also remains unusually low. (See Chart 15.)

Chart 15: WTI & Engineering Investment

Sources: Refinitiv, Capital Economics

All this leads us to expect quarterly GDP growth to slow below its long-run potential in 2023, but remain at 1% annualised or higher. (See Chart 16.)

Chart 16: Capital Economics’ GDP forecasts
(%q/q annualised)

Sources: Refinitiv, Capital Economics

What if house prices fall by more than we expect?

The key risk to our GDP growth forecasts is that house prices could fall by far more than we assume. Perhaps the biggest risk to house prices is simply that they entered this tightening cycle at a much higher level relative to fundamentals than previous ones. (See Chart 17.)

Chart 17: Housing Affordability ($, 000)

Sources: Refinitiv, Capital Economics

This has been driven in part by a sharp rise in investor demand (see Chart 18), with many using the increase in their equity from existing property investments to fund new purchases and ultimately push prices beyond a level that many first-time buyers could afford. If investor demand falls by more than we expect as interest rates rise, prices would need to drop back to a level at which first-time buyers could re-enter the market – at much higher mortgage rates – to balance supply and demand.

Our big concern is that the relationship between house price declines and the impact on the economy is unlikely to be linear. That is, while the economy could largely take a 10% correction in house prices in its stride, a steeper fall risks setting off a negative feedback loop of forced sales and lower prices.

Chart 18: Change in Share of House Purchases by Buyer Type (annual change, %-points)

Sources: Bank of Canada, Capital Economics

While the structure of variable rate mortgages – with most having fixed payments for an initial term – reduces the risk of forced selling due to higher interest rates, history has also shown that economies with elevated debt levels, like Canada (see Chart 19), are more prone to “unknown unknowns”, with some unexpected event triggering a reduction in borrowing and a downturn in activity.

Chart 19: Private-Sector Debt (% of GDP)

Sources: Refinitiv, Capital Economics

These events are impossible to predict, but one potential risk is the recent increase in reports of highly leveraged developers being unable to complete their projects, and pre-construction buyers losing money as a result. Such failures raise the risk that the affected parties will have to sell other properties, adding to the downward pressure on prices, and also increase the chance that demand for pre-construction projects will tumble by even more than we anticipate.

If prices fell by, say, 25% – which would still leave them 15% above the pre-pandemic level – then the downturn in residential investment would likely be even steeper than we currently assume and there would be a much sharper tightening of broader financial conditions as well. A 25% fall in house prices would inevitably cause weakness in the equity market, due to the impact on real estate and financial companies’ share prices, and lead to a tightening of credit conditions as banks curtailed lending and credit spreads widened. In that scenario, the broad tightening of financial conditions would likely overwhelm otherwise positive fundamentals elsewhere in the economy, and ultimately lead to a recession.

Policy implications

Even if that scenario is avoided, the slump in home sales in April reinforces our view that the Bank of Canada is underestimating the sensitivity of residential investment to higher interest rates. Despite its pledge to raise rates to at least 2.5%, the Bank expects residential investment to fall by just 5% this year and 2.5% in 2023, far more modest declines than we anticipate. (See Chart 20.)

Chart 20: Residential Investment (% y/y)

Sources: Refinitiv, Capital Economics

With residential investment already weakening, we remain convinced that the market-implied terminal policy rate, of 3.25% in 2023, is too aggressive. We expect the Bank to pause its tightening cycle once the policy rate reaches 2.5% later this year. If house prices were to fall by much more than we expect, the Bank would loosen policy in 2023, but we doubt it would cut interest rates all the way back to 0.25% for fear of reigniting the housing cycle all over again.


Stephen Brown, Senior Canada Economist, stephen.brown@capitaleconomics.com

Download
Save
Stephen Brown Senior Canada Economist
Continue reading

More from Canada

Canada Chart Book

Tight labour market pushing up wages

The tight labour market is putting upward pressure on wages, with the Labour Force Survey showing a strong 1.0% m/m rise in average hourly earnings in May. While the annual rate of wage growth, at 3.9% y/y, remained lower than before the pandemic, base effects mean it is likely to accelerate to closer to 6% over the second half of 2022. All this adds to the pressure for the Bank of Canada to rapidly tighten policy, particularly as the current weak rate of productivity growth suggests that there is minimal scope for businesses to absorb higher wage costs. We expect the Bank to raise its policy rate by 75 bp in July and follow that with a further 50 bp hike in September and a 25 bp hike in October, to take the policy rate to 3.0%. By then, we think the weakness in the housing market and a sharp slowdown in employment growth will be enough to cause the Bank to pause its tightening cycle, and ultimately remain on the side lines as GDP growth slows below its long-run potential in 2023.

29 June 2022

Canada Economics Weekly

Further rise in inflation cements 75 bp hike

The May CPI data showed that inflationary pressures remain acute across the economy, which reinforces our view that despite growing concerns about the housing market, the Bank of Canada will become a member of the “75bp club” in July.

24 June 2022

Canada Data Response

Consumer Prices (May)

The jump in CPI inflation to 7.7% in May all but guarantees that the Bank of Canada will raise interest rates by a larger 75 bp next month, particularly with inflation on track to accelerate to over 8% in June.

22 June 2022

More from Stephen Brown

Canada Economics Weekly

Cracks in housing market risk turning into chasms

The slump in home sales in April, combined with the unexpected fall in hours worked, point to downside risks to GDP last month. While hours worked are set to rebound in May, there is likely to be much more weakness to come for the housing market. Canada Drop-In (10th May, 12:00 EDT/ 17:00 BST): We’ll be discussing the impact of monetary tightening on the economy and whether the Bank of Canada is risking a policy mistake in this special 20-minute briefing. Register now

6 May 2022

Canada Data Response

Labour Force Survey (Apr.)

The sharp slowdown in employment growth and large decline in hours worked in April present downside risks to GDP growth but, as temporary factors seem to explain much of the weakness, there is likely to be some bounce back this month. Canada Drop-In (10th May, 12:00 EDT/ 17:00 BST): We’ll be discussing the impact of monetary tightening on the economy and whether the Bank of Canada is risking a policy mistake in this special 20-minute briefing. Register now

6 May 2022

Canada Economics Update

Spring market brings deep freeze for home sales

The steep declines in home sales across Canada’s major cities reinforces our view that a downturn in residential investment will prevent the Bank of Canada from raising interest rates by as much as markets are now pricing in, particularly if – as looks increasingly likely – house prices fall later this year.

Canada Drop-In (10th May, 12:00 EDT/ 17:00 BST): We’ll be discussing the impact of monetary tightening on the economy and whether the Bank of Canada is risking a policy mistake in this special 20-minute briefing. Register now

4 May 2022
↑ Back to top