The Canadian mortgage market: it's not just about insurance!
Published: 19 February 2013 | Author: Bernard Clarke
With Mark Carney shortly to become governor of a much more powerful Bank of England, all eyes are turning to Canada. But all most of us know about the country is that mortgage insurance (MI) is a key pillar of its mortgage market.
So, here we take an in-depth look at Canada’s housing finance market, and ask what lessons can be drawn for the UK.
It is really hard to compare different national housing finance systems, and we are certainly not trying to suggest that the Canadian system, or parts of it, are inherently better or worse than the UK model. In many ways, the institutional arrangements of the two countries are "chalk and cheese."
There is evidence that Canada’s compulsory mortgage insurance model, paid for by the consumer, has helped to shelter the country’s lenders from the full adverse impact of the global credit crunch. But of central importance is the activist role played by the Canadian authorities over several decades, in supporting funding markets and providing explicit sovereign guarantees.
So, what are the lessons for the UK?
- First, mortgage insurance comes in several flavours, has an institutional context, and is unlikely to be a panacea for the UK market, at least in the short-term.
- Second, we should be willing to reflect on what market structures might contribute to greater market stability in the UK through the economic cycle, and what role the government and lenders might have in developing more robust housing finance models.
- Finally, we should keep a watchful eye on how well the Canadian authorities manage to cool what has looked like an over-heated housing market, as this may have ramifications for how the UK’s financial policy committee, headed by Mark Carney, chooses to apply macro-prudential policy to mortgage lenders here in the UK.
The Canadian system
The Canadian housing finance system differs from the UK in several key respects:
- Canada has a large public mortgage agency, the Canada Mortgage and Housing Corporation (CMHC), which originates and carries out most housing policy initiatives.
- Major lenders offering high loan-to-value (LTV) mortgages – defined as above 80% LTV since 2007, and above 75% prior to that – must reduce their risk exposure by insuring such loans against default, either through the CMHC or a private mortgage insurer.
- The federal government actively supports securitisation activities.
These three aspects appear to be closely interwoven. Taken together, they enable the Canadian authorities directly to influence housing and mortgage market behaviour to a degree that that does not exist here in the UK. Even if we wanted to, it is doubtful whether we could swiftly replicate elements of the Canadian system. As in the US, some of Canada’s institutional framework and rationale dates back to the depression era of the 1930s.
The role of mortgage insurance
One of the aims of mortgage insurance (MI), since it was introduced in the mid-1950s, has been to make home-ownership possible for a wider spectrum of households.
To all intents and purposes, MI is compulsory on higher LTV business – currently defined as greater than 80% – reflecting federal and provincial regulations on lenders that provide the lion’s share of mortgage lending.
It is worth noting that Canada operates a rather stringent version of MI, even compared to other jurisdictions that have developed MI markets:
- insurance is, in effect, mandatory
- it covers the lender for 100% of the mortgage, not just the excess over 80% LTV
- cover lasts for the duration of the mortgage, not simply the period for which there is a higher LTV; and
- the federal government provides a catastrophic guarantee to back up the CMHC or private mortgage insurers.
In Canada, MI enables house purchasers to borrow readily at higher LTVs, and at interest rates that are comparable to conventional mortgages (that is, those not requiring MI). But those borrowers pay for this, in the form of a one-off up-front fee, which, in most cases, is added to the mortgage principal and amortised over the life of the loan.
There is a banded structure for MI premiums, based principally on the LTV of the underlying mortgage, designed to address the higher credit risk associated with higher LTV loans. Currently, the typical cost ranges from 1.75% to 2.75% of the advance.
Canada's "gold-plated” approach to MI does mean that firms can substantively mitigate credit risk and, with the additional back-stop of the sovereign guarantee that sits behind mortgage insurers, benefit from material capital relief (see Annex B of the recent Basel consultative paper on MI).
Importantly, this set-up appears to have contributed to a less dramatic response on the part of Canadian lenders to the global disruption to funding markets than we saw, for example, from their UK counterparts.
Even if we wanted to, it is doubtful we could swiftly replicate the Canadian systemWhile the UK experienced a dramatic reduction in risk appetite – characterised by the near disappearance of above 90% LTV lending, a quantum jump in deposit requirements and the halving of house purchase transactions – there is no Canadian parallel.
A breakdown of mortgage lending by LTVs is not routinely available for Canada, but a recent Canadian Association of Accredited Mortgage Professionals (CAAMP)/Maritz survey tells us that for home purchases since 2010, 55% of buyers had LTV ratios of 80% or higher. The equivalent UK proportion would be nearer 30%, down from about a half before the credit crunch.
We need to be careful about interpreting these figures, but they are consistent with the notion that Canadian appetite to lend at higher LTVs has been greater and less volatile over the economic cycle.
Government role in funding markets
Understandably, commentators pay considerable attention to the important role played by MI. But this does not tell the whole story.
Even a brief review of the Canadian market suggests that the role of the authorities in underpinning securitisation activities has also been an essential part of the mix.
Although retail deposits have been the long-standing primary source of mortgage funding, securitisation has become progressively more important, just as in the UK.
However, a key difference from the UK is that in Canada activity is dominated by two public securitisation programmes. The CMHC effectively kick-started the use of mortgage-backed securities (MBS), when it launched its National Housing Act Mortgage-Backed Securities (NHA MBS) programme in 1986.
In essence, the CMHC provides an unconditional (and sovereign-backed) guarantee of timely payments of interest and capital, and any capital prepayments, to investors in MBS pools of insured mortgages.
Alongside NHA MBS, the Canadian authorities have also facilitated an efficient and highly liquid Canada mortgage bond (CMB) instrument. Launched in 2001, the proceeds from CMB sales are used exclusively to purchase NHA MBS. The cash flows associated with mortgage pools are restructured into ones that resemble government bonds, paying regular interest payments and repayment of the principal amount at a fixed maturity date. CMBs carry little risk, because of the CMHC’s timely payment guarantee, the absence of prepayment risk and the underlying insurance on all loans in the pool. As a result, they are attractive to a broad range of investors and have facilitated funding supply.
Chart One: Annual issuance of NHA MBS and CMBs, C$ billion
As Chart One shows, NHA MBS issuance grew rapidly during the recent global financial crisis, and especially in 2008-10 when the Canadian authorities implemented a temporary funding initiative, the insured mortgage purchase program. This was similar in some respects to the Bank of England’s special liquidity scheme, but did not increase the government’s risk exposure because all the loans were insured, and the government already carried contingent liability as a result of its backstop guarantee for MI.
It is worth noting too that, as in the UK, covered bonds have also become an increasingly significant funding source for Canadian lenders, with developments closely mirroring those in the UK.
The activist approach of the Canadian authorities towards securitisation activities over many years contrasts with a more ambivalent (and sometimes adversarial) attitude in the UK and Europe.
Although we are not presenting a detailed comparison of national funding models, it seems clear that Canada’s public securitisation programmes have contributed positively to an efficient, competitive and stable housing finance market.
How our markets compare
Given the historical difficulties experienced in the UK housing market, commentators are understandably attracted towards a "grass is greener" mindset. However, the evidence does not immediately support the view that the Canadian model is necessarily superior to our own.
Chart Two: Home-ownership trends, % of households
Recent home ownership trends in the two countries have diverged significantly (see Chart Two), reflecting a wide range of domestic factors. Following a long period of stability, Canadian home ownership rates have been on a strong upward trajectory since the mid-1990s – at least until very recently – such that in recent years UK rates have once again lagged behind those in Canada.
The different trends in the two countries are at their most stark in the post-credit crunch period.
Some of this difference is likely to be explained by the gentler and shorter recession in Canada. Canada undoubtedly weathered the global recession of 2008-2009 better than most countries in the developed world, including the UK, and has emerged relatively strongly, thanks in part to robust commodity price growth.
As elsewhere, the Canadian authorities have opted for substantial monetary easing and other stimulus measures, but a number of these have been centred specifically on its housing or mortgage markets.
In addition, the federal government had relaxed regulations for insured mortgages in 2006, allowing repayment periods of up to 40 years for house purchasers and 100% LTV mortgages (although stringent underwriting standards effectively restricted the latter to higher income professions).
By this time, Canada had already experienced a decade-long housing boom, driven by strong demographics, rising employment and real wages, and declining interest rates (much the same story as in the UK). In Canada’s case, a relaxation of its MI rules boosted the ability of households to borrow, and helped to sustain its housing boom and credit growth (see Chart Three), albeit with less ferocity than before.
Chart Three: Residential mortgage balances, % growth year-on-year
In recent years, Canadian mortgage lending has continued to grow relatively strongly, and the growth in house prices has continued to outpace that of incomes. Canada now has one of the most expensive housing markets in the world, according to The Economist. There are also what appear to be credible concerns about the speculative nature of house-building in some urban areas.
Chart Four: Trends in household debt to income ratios, %
As Chart Four shows, the ratio of household debt to disposable income is now on a par with recent UK and US levels.
Currently, there is little evidence of material stress, thanks to favourable labour market conditions and the wide availability of credit on favourable terms.
The rate of mortgages in arrears in Canada – defined as more than three months of payments not received – has averaged just above 0.3% for more than a decade, and has not been over 0.5% during this period. This is less than one-third of the UK equivalent. And, as in the UK, recent short-term trends have actually been positive. The latest figure shows a reduction in the proportion of mortgages in arrears of three months or more, at 0.32% last October.
Nevertheless, Canadian policy-makers and others have become more concerned about the vulnerability of households to adverse shocks, and especially to higher interest rates. While there has been a noticeable switch back towards fixed rate products, a third of Canadian household debt continues to be on variable rate. The Bank of Canada identifies the most important domestic risk to financial stability in Canada as stemming from the elevated level of household indebtedness and stretched valuations in some segments of the housing market.
In Canada, mortgage debt comprises roughly 70% of total household debt. This contrasts with a figure of 89% in the UK. However, it is worth noting that Canada has also witnessed greater use of home equity lines of credit (HELOCs), which do not get reported in its mortgage data. The Canadian Association of Accredited Mortgage Professionals (CAAMP) recently estimated that there were 2.1 million of these, albeit with about a quarter having no balance outstanding. HELOCs represent something like another 5% of household debt.
When he gave evidence recently to the Treasury select committee, Mark Carney referred to the elevated nature of household debt in Canada and highlighted that since 2008 the federal government has taken a series of steps to progressively tighten MI eligibility rules and mortgage underwriting guidelines.
These are summarised in Figure One.
Figure One: Key recent changes in Canadian mortgage insurance rules
|Announcement date||9 July||16 February||17 January||21 June|
|Implementation date||15 October||19 April||18 March||9 July|
|Maximum amortization period||From: 40 to 35 years||From: 35 to 30 years||From: 30 to 25 years|
|Loan-to-value (LTV) limit for new mortgages||From: 100% to 95%|
|LTV limit for mortgage refinancing||From: 95% to 90%||From 90% to 85%||From: 85% to 80%|
|LTV limit for investment properties||From 95% to 80%|
|Debt-service criteria||Total-debt-service (TDS) ratio capped at 45%||Required that all borrowers quality for their mortgage amount using the greater of the contract rate or the interest rate for a 5-year fixed-rate mortgage in the case of variable-rate mortgages or mortgages with terms less than 5 years||Gross-debt-service (GDS) ratio capped at 39% and TDS ratio at 44%|
|Other selected changes||(l) Established a requirement for a consistent minumum credit score, with limited exceptions
(ll) Strengthened loan documentation standards to ensure reasonableness of property value and of the borrower's sources and level of income
|As at 18 April 2011, mortgage insurance is no longer available for non-amortizing home-equity lines of credit||Mortgage insurance limited to homes with a purchase price less than $1 million|
While all these measures have helped to slow household credit growth, the restrictions introduced in mid-2012 appear to be having the most pronounced impact. The CAAMP has estimated that one in six higher LTV loans that took place in 2010 would not be feasible under the revised criteria.
The critical driver appears to be shortening the maximum repayment period from 30 to 25 years (that is, back to its pre-2006 position). However, because this change was only made in June 2012, nearly a third of loans for house purchase advanced during the year had amortization periods longer than 25 years (a broadly comparable figure to the UK).
A recent Reuters survey signposted that Canada would experience a cooling housing market, with a reversal in house prices, sales and housing starts over the next few years. Recent market indicators describe a noticeable slowdown. While the majority view is for a soft landing, some commentators anticipate a more dramatic outturn.
Lessons or challenges for the UK
It is notoriously difficult to compare different national housing finance systems. Both the UK and Canada have experienced protracted housing and mortgage lending booms, largely on the back of wider economic fundamentals. Recent experience has diverged, and we do not yet know how the Canadian housing boom will end.
One differentiating aspect which clearly stands out, however, is the more activist role played by the Canadian authorities over several decades. Since the credit crunch, Canada’s different institutional framework has furnished the federal authorities with a greater array of policy levers, which rely on stimulating the housing market to buoy the wider economy.
A greater focus on supporting funding markets represents a key policy difference between the two countries, and one that appears to have helped to shelter the Canadian housing market from some of the volatility arising from the global credit crunch.
While compulsory MI for higher LTV lending has sheltered Canadian financial institutions from adverse market conditions, it has not prevented a significant housing boom nor a material build-up in household debt levels. It is clear that MI has not represented a policy panacea in this regard, given that Canadian and UK households today face similar vulnerabilities.
The Canadian authorities have shown themselves willing to intervene directly and repeatedly in an effort to moderate household behaviour. But, at this stage, it is too soon to judge whether Canada can engineer a soft landing for its housing market.
Whatever the outcome, there may be some useful lessons for macro-prudential regulators in the UK and elsewhere on the potential for fine-tuning housing market developments.