New US Export to Canada: Deleveraging

This Time It’s Different is the title of a 2009 book by economists Carmen M. Reinhart and Kenneth S. Rogoff that details “Eight Centuries of Financial Folly”. The phrase “This time it’s different” connotes the similar belief held by investors during every mania that eventually went bust.

Every financial crisis shares similar eerily similar characteristics. The mania- whether it refers to 17th century Dutch tulips or 2006 Las Vegas condos- begins with an incorrect pre-judgement that prevailing conditions will never worsen, and investors are missing out on a new stellar opportunity. When the catchphrase “this time it’s different” becomes the rationale behind the investment scheme, it’s never long before the underlying asset comes crashing back to reality.

Currently, there are signs that Canadian housing buyers are proclaiming, “This time it’s different”. Household debt-to-income ratios have risen over 160% and the amount of construction in the major cities indirectly accounts for 30% of economic activity, both at record levels.

(Disclaimer: Canadians are some of the most genuine, authentic folks on the planet whose zealous love for country is relatively unmatched even though much of the country spends half the year in freezing temps and complete darkness. Canada is the only other country that participates in US professional sports- hockey, baseball, and basketball. For my readers north of the border, please don’t by offended by the following critique, it’s written with objective concern and hope that the Canadian government acts early and aggressively to halt the next potential financial crisis. I also intend to shed light on the common misconception that housing bubbles in the US were caused by the deregulation of the financial sector, the widespread use of derivatives, and government backed loans. When you see what’s happened in Canada since 2008, you might reconsider the widely accepted reasons for the US housing bubble.)

The aftershocks of the 2008 financial crisis continue to reverberate around the globe, and the varied central bank responses have painted the backdrop for extremely volatile asset and currency price swings. As US consumption behavior changes, it impacts consumption elsewhere.

While the US household has gone through a deleveraging, Canada has witnessed the opposite:

Beginning in 2008, US households have increased savings and paid down debt. On the other hand, Canadian households have added debt rapidly, making the US look thrifty.

With an overleveraged Canadian household, the drop in oil and subsequent capex cuts by Canadian drillers could create a Minsky moment- the point in the credit cycle where asset values (re:housing prices) suddenly drop because the lengthy period of prosperity and increasing value of houses brought out speculators with borrowed money, who now attempt to sell simultaneously.

Before discussing exactly how Canadian debt started mounting, here are a few accounting identities for open economies:

1. The current account and the capital account must always balance to 0. For instance, every $ that arrives in Canada as payment for goods must be used to buy something from abroad, as a foreign payment, or the Canadian merchant can save the $ by purchasing a US asset.
2. A current account surplus or deficit is equal to the difference between total domestic savings and total domestic investment. A surplus in the current account (from excess domestic savings) must be exported abroad. It follows that exporting capital is the same thing as importing demand. In other words, Canadian savers balance out with US consumers, and any changes affect the other one.
3. All production must be either consumed or saved. (If total production = GDP; then Savings = GDP — Consumption.)

Because of these accounting identities, all Canadian production must be either consumed domestically, invested domestically, or exported to bring in foreign demand. The reverse is also true.

As the destination for 95% of Canadian exports, Canada’s trade account is inextricably linked with the US. Any change in the difference between US consumption and savings must be followed by an opposite change in Canadian consumption and savings. If production stays constant and Americans consume less, the rest of the world (especially Canadians) must consume more.

Thus, the great Canadian credit expansion is underpinned by the increased savings rate of the American consumer.

Between 2000–2008, the US savings rate averaged 2–5%. The financial crisis led to a jump in the US savings rate to a range of 4–6%.

And since any increase in US savings must equal an increase in consumption somewhere else, an increase in US savings caused the Canada trade balance to swing from a surplus to a deficit. Canada was no longer able to import US demand. This is clear in the following chart:

Prior to the financial crisis, Canada ran trade surpluses, however, the Canadian balance of trade fell to deficits concurrent with the jump in US savings.

Once Canada became a net importer, there are a number of ways that US surplus could have been absorbed:

1. The Canadian investment rate rises above the savings rate enough to match Canadian deficit with US surplus. This was unlikely as the world was stumbling through a global recession, and investments did not appear productive.
2. Canada imports US capital at low or even negative real rates which finance a real estate boom that forces up Canadian investment to absorb US surplus. ← — — Usually what happens.
3. Canada consumption rises faster than GDP (on the back of cheap US capital exports).
4. Canada could have refused to absorb the excess US savings by raising taxes and cutting fiscal spending, causing a rise in domestic employment. The higher unemployment would have brought down the Canadian savings rate as production fell faster than consumption. But what government would purposely cause unemployment?
5. Devaluation of the loonie would have lowered effective labor costs and forced Canadian savings back on the US as exports increased. However, Canada seemed to have picked up “Dutch disease” (covered later) as oil’s rise above $80 sparked a gold rush in the oil sands.

It appears that the forced consumption in Canada was manifested in a housing boom:

Although they rose in tandem throughout the 00’s, Canada housing prices disconnected when US housing prices started falling in 2008. Canada likely avoided a drop because a sudden increase in US household savings provided low or negative real financing costs through the current account.

Leading into 2009, there was a small dip but nothing equivalent to the 30% pullback witnessed in the US market. The Canadian mortgage origination market greatly differs from the US, and this could have insulated their housing market from the US collapse.

For starters, the Canadian system favors the creditors. In the case of a default, the bank can prosecute the debtor and his/her other assets, and not just claim the house. On the contrary, the US has onerous foreclosure proceedings where the bank only has the house as recourse.

In addition, the US standard 30-year mortgage doesn’t exist in Canada, where the typical mortgage is a 5-year fixed amortized over 25 years. Canadian borrowers get a rate reset every 5 years, leaving them exposed to a sudden rate increase. There are also high prepayment penalties that prevent the Canadian borrower from paying down the loans, favorable to the bank.

The largest difference is the absence of mortgage securitization in Canada. Banks and underwriting terms are heavily regulated by the government. Canadian banks must keep loans on their books and maintain “skin in the game”. This largely avoided the “creative” mortgage structures that defined the US housing bubble.

Finally, in Canada, mortgage interest isn’t tax deductible, taking away any incentives by consumers to over-borrow as a tax reduction strategy.

One similarity is the existence of government-sponsored enterprises that provide a backstop for home loans. Similar to Freddie Mac and Fannie Mae, the Canada Mortgage and Housing Corp (CMHC) accounts for 70% of all mortgage insurance and mandatory on loans covering less than 80% of the home value.

These differences would lead an investor to think of the Canadian mortgage market as “safer” than the US. Remember, there have been housing crises since the Roman Empire that didn’t originate in subprime or adjustable rate mortgages. The prevailing wisdom that the Canadian market is somehow protected because they survived 2008 relatively unscathed underpins Canada’s “this time it’s different”. This permissive thinking allows for the misperception of risk. Namely, the perception of stability fosters instability.

In 2009, Canadian banks compared the low level of defaults (0.45% vs 5% in the US) and credit expanded dramatically. Here’s the chart that shows the exponential growth in CMHC insurance over the past 5 years which the government was forced to cap in 2012:

The first signs of stress in housing showed up in Calgary in December 2014. Home sales fell 7.5% from the previous year, and new listings surged by 42%. Calgary is the epicenter of the oil sands region, home to many multinational energy companies now facing unprecedented capex spending cuts.

Last week, the Canadian Association of Oilwell Drilling Contractors predicted a 41% drop in active drilling rigs, and estimated a loss of 22,000 jobs. The network effect of those job losses that will be significant as architects, builders, servers, lawyers, financial planners, etc are linked to the Canadian energy growth engine. When the energy jobs start to disappear, so do the businesses they support. The same rule applies to finance in NYC and entertainment in LA.

Some might think that the scope of this problem will be limited to the energy-dependent Alberta region, and that cities such as Vancouver, Toronto, and Montreal will be unaffected. This might be true if the rise in housing prices had not been a national phenomenon:

It appears that Canada’s 2008 bubble was rescued by an increase in US savings that fueled even more Canadian consumption and pushed the housing index’s eventual mean reversal further into the future. This process has increased leverage in the system as credit increased and prices kept climbing.

A pullback in Calgary house prices can have a detrimental effect on the rest of Canada. In the US, up until 2008, the Fed believed that poor performance of subprime (which accounted for 20% of the mortgage market) would be an isolated event. Very few predicted that credit impairment in subprime would lead to significant tightening of overall credit. A pullback in prices in one area leads to a similar pullback in another. If the same banks operating in Calgary are also lending in Vancouver, Toronto, and Montreal- why should this episode be any different?

Signs of financial stress are already appearing. The Canadian $ has weakened almost 20% in the past 6 months:

Typically, a currency decline should be beneficial to jobs as the exports suddenly become more attractive at lower prices. However, persistently high oil deferred resources from manufacturing tradeable export products to digging holes in the ground in search of oil. The following chart illustrates this shift:

Thus, a devaluation will not be enough to replace the economic growth lost by Capex spending in the oil sands. Workers need to be retrained and hired into other areas of manufacturing. In the US experience, depressed housing prices immobilized workers. A factory worker in North Carolina, underwater on his mortgage and not wanting to sell for a loss, could not move to North Dakota or Texas where energy jobs were abundant in 2010. Canada will face the same labor mobility problems as energy jobs are lost in the west and manufacturing jobs are in the east.

Canada may have suffered from “Dutch disease”, a term used to describe the problems faced by Dutch manufacturers after the discovery of natural gas in the 1970s. Because of this newly discovered resource, foreign demand for Dutch nat gas exports drove up the value its currency (the gilder) relative to the value of its trading partners. The higher currency values caused other Dutch exports to become less competitive in international markets. Economists point to this syndrome as a reason why resource poor countries such as Japan, Hong Kong, and Western Europe have outpaced oil-rich countries such as Nigeria, Venezuela, and Russia.

While oil sands have been mined for over 100 years, the vast majority of production isn’t economical under $80, due to the initial capital outlays to dig the mine. For the last 10 years, the loonie has been closely correlated to the price of oil. A high oil price means strong export demand for Canadian oil, which strengthens the loonie. The loonie dropped when oil traded $30 in 2008, and more recently when the price of oil was cut in half:

The recent drop in oil and the C$ has been positive for non-energy exports as highlighted in a recent note from Krishen Rangasamy, senior economist at National Bank of Canada:

“The boost provided by the uptick in U.S. demand and the weaker Canadian dollar is being felt across the entire manufacturing sector… Goods with a strong U.S. export component had a particularly strong month. Motor vehicle sales jumped 9 per cent, while vehicle parts were up 4.9 per cent. Machinery sales gained 5.2 per cent. Wood products rose 4.3 per cent and paper manufacturing gained 4.2 per cent.”

It remains to be seen if the increase in exports can positively impact real wages as the weaker C$ also means higher import inflation. Without continued increases in real wages, those elevated debt-to-income ratios will contract and bring down home prices. These actions may perpetuate a vicious cycle where the falling price of homes brings out more sellers, not buyers. And we will again learn, this time is not different!