Duncan Maclennan and David Graham
9th October, 2017
- Pressured Markets and Policy Measures: Canadian Responses.
Shaping Futures: Canada in Context
The University of Glasgow has recently led a major cross-national collaboration, Shaping Futures, that explored emerging issues in housing policy. The project involved the Universities of Glasgow, Toronto and New South Wales, along with non-profits, government agencies and charitable trusts drawn from Australia, Canada and the UK. The partners have undertaken an 18-month long knowledge co-production and exchange that has assessed changing housing systems, challenges in private rental provision, the evolution of non-profits, the governance of housing and the pressures on growing metropolitan housing markets. Although there were significant differences in macroeconomic background and policy circumstances across the three countries there were also important commonalities. Estimated needs for non-profit housing provision were rising sharply, homelessness was rising and in the market sector long periods of price pressure, particularly in growing metropolitan areas, had seen prices and rents rise well ahead of incomes. This has resulted in a broad affordability difficulty across large parts of the income distribution. Younger households, rationed into rental housing by rising prices and ‘flat’ labour market entry wages, faced increasing competition from baby-boomers buying homes to let and as price pressures prevailed and there was a concern, supported by little evidence, that foreign buyers were driving markets to new levels of un-affordability. More recently, as price booms peaked and are now stalling, financial stability and household debt exposure are attracting financial stability policy concerns.
This paper has evolved out of that collaboration and is concerned with recent Canadian policy shifts to cope with, and remove, house price pressures in major metropolitan areas, Vancouver and Toronto.
Canadian Housing Markets: Affordability, Productivity and Stability Difficulties
Whilst many Canadians are, by any standards, well housed there is a growing unease with national housing outcomes and how they affect Canada’s big goals for the environment, the society and the economy. The environmental footprint of Canada’s residential development remains relatively ‘dirty’ (Harcourt, 2007). Housing wealth disparities and housing system segregation processes are shaping more unequal wealth distributions and geographies in Canada (Hulchanski, 2015). This short paper, however, concentrates on major economic concerns arising in, and impacting, the housing market sector. These include inequality, productivity and stability effects.
Housing market outcomes, and rising rents and prices have contributed to rising inequalities of both income and wealth (and these have impacted key social goals for mixed communities, reduced segregation and increased social mobility) (Maclennan and Miao, 2017). There is now a widely recognized dual crises in Canada, as in the UK and Australia. There are burgeoning estimated needs for provision in the affordable or non-profit housing sector. Much more widely, there is a growing crisis of affordability for younger Canadians who with prices rising ahead of incomes are finding it increasingly difficult either to enter the housing market or to save as quickly as house prices rise. Paying for housing, more specifically the capacity to make required equity downpayments, is now a challenge not just for poor but for middle income households.
These inequality and affordability outcomes often attract interest for reasons of intra and inter-generational fairness. Baby-boomer Canadian parents and grandparents are increasingly aware that their children are not financially able to form permanent relationships, live independently and buy homes at the age, and even life-cycle stage, that they had achieved such goals. However, the consequences are not solely about ‘fairness’. Many of these outcomes are also directly associated with reducing productivity performance in the Canadian economy, not least by diverting disposable incomes from consumption and other forms of investment into either existing bricks and mortar (with no productivity gain) or relatively low productivity new construction.
The weight of these effects has grown as the favourable terms of trade that drove Canadian income growth for much of the last quarter century have deteriorated (hopefully temporarily). These, and related productivity questions, have attracted little policy or academic interest in Canada. The economic policy debate is devoid of discussion of productivity effects of housing outcomes and, although there is much discussion of an anticipated housing strategy to address growing inequalities, current policy actions at both federal and provincial levels have been primarily driven by the different goal of maintaining economic and financial stability. Rising house prices in some metropolitan areas, exceeding 12pc in Vancouver in 2015 and 20pc in Toronto from April 2016-17, have been associated with high and still rising national household debt to income ratios.
These sharply rising prices have prompted two policy reactions. First, the Federal government moved first to combat what it saw as a potential financial instability emerging. At the end of December 2015, it announced that from February 2016 increased downpayments would be required on homes valued between $450,000 and $1million. It followed these actions, in December 2016, with measures to raise the quality of mortgages attracting subsidised mortgage insurance as well as closing tax loopholes for non-residents claiming Canadian homes as their principle residence. Further measures to stress-test uninsured mortgages have been proposed in the summer of 2017 with a view to their implementation in late fall 2017. And, after a long period of low interest rates the Bank of Canada raised the base interest rate from 0.5 to 0.75 in July and then to 1pc in September 2017. Taken together, this is a series of modest measures that may constitute a significant ‘leaning against’ housing market activity.
The second set of reactions have been at provincial levels. British Columbia undertook a series of small legislative reforms through 2015 and early 2016 that reshaped the property tax exemptions for non-residents and introduced measure to reduce fast ‘flipping’ in rising markets. More significantly, in July 2016 the Government of BC announced the introduction of a 15pc foreign buyers tax for the Vancouver region. The Government of Ontario followed suit in April 2017 announcing a 15pc foreign buyers tax in the Greater Golden Horseshoe region as part of a wider set of measures to improve housing affordability in Ontario.
Governments are rightly concerned about housing and mortgage market stability. The combination of what financial authorities in Ottawa now perceive as some relatively poor quality and relatively risky lending in the post GFC stimulus programme (see further below) with subsequent sustained house price rises in major markets provide a clear basis for stability concerns. This paper, whilst not questioning the direction of travel of financial policy for housing market stability does, however, raise several questions about the coherence and coverage of present Canadian policy approaches. Has the financial stability concern been exaggerated and raised to ‘national’ significance (see Bank of Canada, 2016) when in reality is it a highly-localized problem? Has the financial stability goal been pursued by one part of government without regard for the wider housing system effects it creates? That is, has there been a failure in government and governance structures to link financial and real sector policies and have ‘housing’ concerns fallen outside the key decision taking arenas. The concerns raised below include whether there is: attention to evidence in decision taking; linkage in financial and housing policy decision taking at federal levels; any effective coordination of provincial and federal actions; a neglect of the importance of local variety in metropolitan and regional housing markets in Canada; a sense that the combined effects of different policy actions may be aiming to do too much too quickly and may needlessly destabilise markets; an absence of consideration of policy effects on the wider goals of productivity and reducing inequality. These different questions add up to a broader concern as to whether there is a coherent housing market policy for Canada, and especially the home-owner sector, being evolved amidst the strategic housing policy measures currently being developed within government.
Housing Market Policies: Missing Links
House price rises and instabilities are major, not minor, issues for modern economies but it is not clear that modern governments are well equipped to deal with them. Housing policies in Canada have been downloaded and downplayed for two decades. At the federal level, major housing policy autonomies have been downloaded from federal government to provinces since the middle of the 1990’s. CMHC had long ceased to have significant housing policy functions a decade before the advent of the Trudeau government though it had key roles in credit market policy after the GFC. Within government small housing scale housing policies, by international standards, have been located within human services departments and have been focused on issues of poverty and disadvantage rather than housing system functioning, supply and stability. At the provincial level, in this millennium, the governance of what is now defined as ‘housing policies’ has become fragmented and marginalized and shrinking housing expenditure policies have, as at federal levels, focused on the homeless and poorer households. In consequence housing ministers/ministries have lost sight of the imperative of managing an efficient housing system that ensures an adequate supply of housing as essential infrastructure for successful economic growth. With a more relaxed approach to allowing ‘demand side’ tax expenditures that assist home owners and private landlords, the lack of attention to housing investment and supply strategies has transformed owner occupied housing from a means of saving to a prime sector for speculation, not just by foreign investors in rental housing, but for the clear majority of Canadian home-owners.
The Federal Government, since 2015, have made a needed and welcome return to both housing policy debate and housing policy action. Provinces and providers have responded to policy interest. There has already been major investment in remedying inadequacies in existing lower income housing, the budget of 2017 increased (and time extended) the budget for Federal support for affordable housing and new policy approaches are being designed with a potential housing allowance being modelled/designed and a commitment made to transform the non-profit provider sector. All these policy initiatives are much needed. However, they do not appear to have the effective, stable functioning of Canadian housing markets as their focus of concern. Canada, in short, needs a new focus on housing market policies that complements the decisive actions on financial sector stability. Given the policy settings for financial stability, what is the consequent housing market strategy that will deliver the key policy goals of federal and provincial governments?
This paper focusses on the market sector, and home-ownership. Home-ownership remains, by far, the majority tenure in most of Canada, and it remains the tenure of aspiration for the clear majority of Canadians. There is now a widespread difficulty of lower income households and younger households with potentially high lifetime incomes in entering ownership. Alongside the well-recognised crisis in Canada’s affordable housing sector there is a much wider crisis of housing affordability for middle income and younger households as well for growing numbers of younger Canadians. It is yet unclear how policy is responding strategically to confront these difficulties. We do not share the view that resolving the problems facing younger households is to further ration them out of ownership by financial policy measures and to simply expand market rental supply. Offering 25 to 40-year-old households with jobs more rental housing whilst they save to become home-owners as house prices rise ahead of income increases is not a credible strategy for an ownership based market or society. We examine the possibilities and critique some of the rather ad hoc policy measures that have been implemented in the last two years.
The next section of the paper (II) outlines our understanding of the complex and cumulative processes of house price rises in Canada in this millennium, section III then considers some consequences of these changes. Short-term policy responses at Federal and Provincial levels since 2014 are then considered, and downsides highlighted, in section IV and the concluding section, V, indicates some broader directions for longer term policy changes.
2. Driving Demands and Reinforcing Changes
A Pro-Housing Market Context
Canada, until 2015, has enjoyed rising incomes for almost two decades as strong terms of trade favoured major Canadian export sectors (oil, minerals and food). The lower global oil price and the reduced growth rates of China mean that income growth now relies more on raising sluggish productivity growth rates in the Canadian economy. Encouragingly, Canada has had a strong growth performance through 2017. The demand for housing is, near universally, income elastic so that households spend proportionatley more on housing as incomes rise. That income driven increase combined with rising population and household numbers in Canada (rising from an annual increase of 1pc at the millenium to 1.3pc by 2015) have seen a steadily, significant rise in annual housing demand. Interest rates also have a significant impact on saving and borrowing for housing. In Canada, as in most of the other OECD economies, there has been a long steady decline of mortgage interest rates from around 7pc in the late 1980’s to 5pc at the end of millenium to 4pc now. The future outlook is that rising interest rates, by modest amounts over a quite protracted policy, will characterise the five years ahead.
Tracking Price Changes
Over the same ‘long boom’ period, with significant cyclical variations, residential investment (including new construction, renovations, existing sales plus tax) grew from 15 to 21pc of GDP between 1981 and 1989, fell back to 5pc by 2001 before rising back to 20pc by the GFC, fluctuating around that level to 2014 and then rising towards 22pc in 2016. As discussed below the value of new housing output was less elastic. In more technical terms, the period has been one of a historically low user cost of housing capital with relatively constant relative tax settings, low interest rates and rising real house prices so that housing investment has flourished whenever households could raise incomes or deposits to leverage their housing holdings. It is also in this period that human capital flows, espcially of skilled workers, have increased across national boundaries as have flows of speculative investment. ‘Global’ reach into ‘local’ housing markets has intensified, in turn sweeping aside some of the endogenous economic relationships that limited upswings and downswings in housing prices.
It is important to keep this ‘big picture’ of favourable, but changing, fundamentals for housing investment in Canada in mind. In the understanding of house price dynamics set out below we consider in more detail how rising housing markets faced with inelastic supply responses may quickly develop reinforcing dynamics of first inflationary expectations, ‘irrational exuberance’ and non-resident speculation. However, these processes are most likely to be pronounced disturbances within particular local or metropolitan markets rather than the ubiquitous, everyday reality of all Canadian housing markets. We make this point both to emphasise the likely importance of ‘fundamentals’ in market development but also to highlight, for further discussion below, the need for booms and emerging bubbles to be contained by local rather than universal policy instruments.
In the UK and Australia regular analysis of the housing market, at national, regional, metropolitan and local levels is facilitated by the regular (monthly) publication of house price series that capture all segments of the market and that are standardised for seasonal and other shifts in the mix of properties traded (and this is an important standarisation as cycles or ripples of housing market activity tend to shift the composition of trades quite systematically over short periods of time). In Canada, MLS systems have been slow to consolidate, and publish Geographic Information System based, standarised price series (leaving Canadian governments and consumers relatively poorly informed of housing market change patterns). That is, qualitative shifts in what is being traded may distort ( and in upswings exaggerate) published MLS measures of price inflation. For example (Globe and Mail 15th September), the public debate about the effect of the new ‘foreign buyers tax’ in Ontario is now informed that average house prices in the Greater Golden Horeshoe area have fallen from $921k to $732k, or a reduction of 20pc, in the space of three months. However as market reports emphasise the shifting composition and geography of stock sold in that period (with trading in higher value sectors falling back quickly relative to the lower-priced sector) it is clear that the 20pc figure has no firm value in market analysis. Governments that cared to understand their housing markets could easily have GIS systems of repeat sales indexed prices in operation if this area of policy debate is to move on from alchemy. Figures from individual banks, that are often standardised, may be conditioned by the sectors and regions in which they operate.
With these caveats in mind, house price data from the Royal Bank of Canada (RBC) show that in the major cities of Canada annual house price inflation rates ran at 5pc at the start of the millenium but from 2002 until the GFC rose through 10pc peaking at 15pc in 2008. After falling back by -5pc in 2009 annual price growth had recovered to 15pc by 2010 and then ameliorated to around 5pc pa until they rose sharply back to 15pc rises in 2015 and 2016 (very similar price cycles were apparent in Australia but with more pronounced peaks and troughs and with higher average nominal inflation rates).
The Canadian cities that led price booms changed within the millennium, with Edmonton and Calgary having huge peaks and falls 2005-2010 with Toronto and Vancouver leading inflation rates post 2012. Looking across not just the 5 major cities but also including the wider set of CMA’s, if the cities driven by fluctuating oil prices are excluded, then there is a very clear relationship between house price rises and economic growth. Figures 1 and 2, below, sourced from Lowe (2017) highlights the very different price experiences of different metropolitan housing markets after 2014-17 with house price inflation rates clearly and positively related to real house price growth rates. For all metropolitan areas, that had price appreciation rates above the average for Canada, different economic growth fundamentals drive different house price outcomes, see Figure 3. The growth-house price inflation chart (3) is drawn from work by the Bank of Canada and they also recognise (2016) growing divergences within Canadian regional and metropolitan housing markets and they suggest a trifurcation (The Oil Province: BC and Ontario: the rest of Canada) of price cycles. Regional differences in cycles are well established in larger, more diverse economies (see Chowdhury and Maclennan, 2015). The Bank of Canada’s own research (BOC, 2016) makes it clear that Canada’s Metropolitan areas have different economic trajectories, different house price appreciation patterns and different local housing market adjustment capacities. The decision of the financial authorities to then apply nationally uniform mortgage regulation restrictions is then somewhat surprising, especially given the absence of any regionally differentiated housing market policy
Source: Bank of Canada
New Economics of Metropolitan Growth
Policy needs to focus on metropolitan housing system differences as well as monetary policy measures. The recent, post millennium, patterns of metropolitan development in Canada share much in common with similar advanced economies. There is a growing, and useful, conventional wisdom emerging in academic and policy debates about the role of major cities in national economic development. The notion of cities as a terrain of physical decay, social disadvantage and economic decline that pervaded policy debate into the 1990’s has passed and that cities, following the research arguments of Glaeser (2009) and others (Puga, 2010) are now seen as key locations for innovation and high level labour market functioning facilitated by agglomeration economies underpinning a ‘new urban economics’. Canadian experience is consistent with this broad view (Conference Bard of Canada, 2016). Whilst supporting this notion it is important not to forget two other important sets of factors that influence the performance of metropolitan areas. Large metropolitan areas are often major points of connection to wider national, and now global, systems so that their global visibility and reach is of a different kind from smaller centres or rural areas. They are much more likely to be connected into the flows of trade, labour and ideas that are rising in the global economy and in consequence to be more evidently impacted by both the growth prospects and negative shocks that arrive from outwith national boundaries.
Connection and agglomeration both matter. Further, both agglomeration and connectivity gains play out within, often, relatively fixed local systems of housing, land and infrastructure systems so that non-tradeable private and public systems may be prone to congestion effects and costs. Albuoy (2012) and others have highlighted that different cities have different productivities in combining labour, land and capital into required housing outputs and that the costs of growth and congestion, largely reflected in growing scarcity (or economic) rents to existing landowners, may attenuate different productive activities. And it is this latter regard that more focus is needed upon housing markets in Canada and their consequences for growth and competitiveness as well as rising wealth inequalities rising from house price increases. This issue is yet hardly discussed in Canada let alone featuring in policy thinking and action. Rising house prices have attracted national policy attention because macroeconomists are concerned that rising borrowing for housing investment, discussed below, may result in increased risks for financial and economic stability.
Against this framework of ideas on metropolitan economic development our concern is that inattention to housing as economy serving infrastructure leads to weak supply responses to growing demands. That raises ‘congestion’ costs of rising house prices and rents, with significant distributional and mobility effects discussed below, but also has direct negative effects on productivity growth. With residents of Canadian cities putting ever greater shares of their income into purchasing primarily second-hand assets as prices rise there is an inevitable reduction in consumption on other goods and services (with higher productivity) and a distortion of investment and savings patterns that may have macroeconomic significance.
To understand these processes better and to identify likely gainers and losers it is important to get below the reductionist perspectives of macroeconomics. Although driving forces may be global, regional or local in origin, the geographically fixed nature of housing, and indeed the pre-dominance of local moves in most markets, means that housing markets, to quote Edwin Fisher’s (1958) description of six decades ago, ‘are ineluctably local’. This means that real housing markets have real space and time dimensions (that macroeconomics, by definition, has to abstract) and will often reveal significant path dependencies in their development. These are not simply academic observations about the right conceptual lens to deploy in answering real housing market questions but they highlight the importance of the local dimensions of preferences, processes and knowledge in effective housing-policy making. The design of policy making processes requires dialogue and cooperation between governments operating, with different instruments, at different geographic scales.
In the now extensive literature on the housing dimensions of the GFC there is a widely- accepted view amongst economists that economic modelling can point the direction of change in economic systems when national economies are developing within periods of near normal ‘punctuated equilibrium’ settings. Once disequilibrium develops forecasting often fails as transformative processes in expectations, crowd behaviour, collective fear take hold but they have no role in the economic theory underlying modelling. It is not enough to identify ‘irrational exuberance’ and suspend analysis until a ‘new normal’ appears. For housing economists and analysts, a key role is to try to understand systematic features of how pressured housing markets behave, the processes in play and reinforcing and dampening mechanisms.
Housing economics, despite the pervasive view in media commentary on housing issues, is not Economics 101. Markets have recurrent features that core microeconomics will usually assume away and good-policymaking will take them into account. There is a school of thought that the price inflation problems observed in large metropolitan areas arises from supply inelasticities that are largely a reflection of land use regulations and planning, for instance the government of New Zealand leans towards this view. We have little doubt that some regulatory regimes slowdown development, whilst we also recognise that planning can reduce developer uncertainties and facilitate coherent infrastructure provision, but it also needs to be recognised that under-investment in infrastructure inhibits housing supply even when zoning is permissive and that developer behaviours, when they are also landowners, and construction sector shortages may also play a part. There is little coherent empirical evidence on the strength of these different supply inhibiting factors for Canadian cities. However, there is sufficient evidence of inherent supply side lags and market failures to suggest that a ‘leave it to the market’ to resolve flies in the face both of recent experience in Canadian markets as well as decades of housing economics research. Federal level imposition of monetary, and fiscal, policy measures with housing objectives or impacts need to be set within coherent measures to address regional diverse supply side pressures. If they are not, then policy outcomes will not be optimal.
The discussions in the shaping Futures Project has identified a series of recursive and reinforcing processes operating in all the major metropolitan markets we examined, including Toronto and Vancouver. Taking these processes into account leads to a very different policy debate, stressing the need to design more stable markets for the longer term, and discarding the ‘foreigners to blame’ rhetoric that has been apparent in Canada (and Australia, and of course Brexit) since the middle of this decade.
The ‘Shaping Futures’ Synthesis.
There is now a system wide problem of housing affordability in major metropolitan housing markets in Canada, Australia and the UK (though by mid-2017 the effects and uncertainties of potential Brexit are dampening housing investment and prices in London). Rising housing costs, largely driven by house price increases above national averages, have had several well-defined effects/stages that are common to the cities that we have been looking at:
Stage 1: Rising Demands for Housing Consumption Drive Initial Price Gains
- Rising house prices in markets with expectations of future population and income growth (the fundamentals for Vancouver and Toronto) do not result in immediate reduced or ‘rationed’ demands for owner-occupied housing but encourage existing non-owner residents to attempt to accelerate their purchases ‘to get on the escalator’, reduce future costs and make ‘gains’.
- Increases in house prices, loan to value ratios and mortgage outgoings for owners (and rents for non-owners) as a share of incomes, immediately diverts household spending from other (higher productivity in production) goods and services and whilst this may increase employment in the housing sector, employment and productivity growth are attenuated in other sectors.
- Housing prices, and land prices, have risen most in locations accessible to major employment locations, especially city centres. In consequence, poorer renters are displaced outwards, poorer home-owners are gradually shifted to the metropolitan edge, where land and housing prices are lower (but travel costs rise significantly) and growing volumes of non-profit land purchase also shifts outwards. There is growing separation of income groups within major metropolitan housing markets, rising spatial concentrations of the poor and increasing income/wealth inequalities reflecting differences in net housing wealth.
- As prices rise existing owners make gains in housing wealth. They may extract some of that to reinforce upswing consumption, reinforcing economic instability in the upswing but that effect is likely to be less significant than the largely unmeasured consumption effects of those facing higher rents and prices; housing wealth is now the largest element of household wealth in all three countries. Broadly it is just under half of household wealth, two thirds of that wealth are held by owners aged over 65 and the proportion acquired by real mortgage repayments appears to be less than 25pc; unearned, and often untaxed, housing capital gains have turned housing ownership from a savings vehicle to a speculative venture. Intra and intergenerational inequalities appear to be rising in Canada as in the other countries.
- Rising house prices make home-ownership, despite the long period of low interest rates post, less affordable and riskier. Loan to value/price ratios have increased sharply and loan to income ratios have also risen significantly although low interest rates have kept repayment burdens, where loans can be obtained, at moderate levels. As younger households who aspire to home-ownership ( and across all three countries these aspiration levels have remained more stable than actual entry rates, with 9 out of 10 households usually aspiring to eventual ownership) run into rising house prices, deposits required etc. they have increasingly entered and remained longer in rental housing; former owners divorcing also constitute a growing stream of older households into renting and older households with the share of owners with active mortgages beyond the age of 65 rising significantly ( this has an over 65 labour supply effect). Larger shares of under 35’s now enter market rental and stay in the tenure longer.
Stage 2: Sustained Real House Price Increases Raise the Investment Demand for Housing
- Once house prices are a relatively safe one-way bet, households with savings and retirement plans recognise that buying a house to let is a rational investment strategy for them: housing shortages drive up rental returns and making letting easier and, at the same time, tax advantages of different kinds lower the user cost of capital to acquire an appreciating asset. Since the GFC monetary policy in all the Shaping Futures countries has sought to keep interest rates low and Canadian rates have been at historically low levels. This policy stance has depressed returns to savers, including those with pension plans. However, households with collateral have been able to borrow and, in consequence, have either switched savings from monetary instruments to purchase real property to let (BTL) or leveraged their existing housing assets to buy additional properties. There is a lack of clear data on these patterns but 50 percent of private sector rents paid now accrue to baby-boomer home owners in the UK and in the major Canadian markets there are estimates that the number of households owning a property to rent has increased by at least 10 per cent since 2010. In all three countries, growth in private rental provision has been an important component of alleviating metropolitan housing shortages since the millennium however BTL investment does not always create an efficiently financed and managed rental housing sector. Further, competition for small rental properties is often fiercest in what had been typical sub-markets for first-time homebuyers and this reinforces price changes against the latter. In many regards, middle-aged metropolitan households have come, in some sense, to exploit each other’s children! With this mechanism in play metropolitan housing markets acquire a new important dynamic and a significant increase in the demand for ‘investment’ properties as opposed to homes to live in. This market dynamic reflects not just tax arrangements and more or less free rental market arrangements but the effects of housing wealth held by older households being leveraged at low mortgage borrowing rates.
- Housing wealth owned by older households may drive market prices against young home-buyers. However, older owners (parents and grandparents) are now using their acquired wealth, even withdrawing their own housing wealth, to transfer chunks of equity to their children/grandchildren. In the UK, some 80pc of first time buyers have a substantial tranche of equity provided by parents and is estimated that UK parents will supply £6bn of such loans/gifts in 2017. But what about large families? What about the daughters and sons of rental sector parents? Social mobility must fall. There is evidence that these processes are now well entrenched in Australia but there is a dearth of relevant research in Canada though anecdotal signs of similar processes.
Stage 3: High and Rising Prices Boost Speculative (and ‘Safe Haven’) Investment Demands.
- Once house prices are rising, expectations of future price increases may increase rather than reduce (equilibrate) ownership demand in subsequent (medium term) periods. BTL bidding further raises prices. And cities with sustained rent and price rises, and this is true of the largest metropolitan areas, may then come to be regarded as ‘safe havens’, for shelter and money, for households living in less economically and politically stable countries. And of course, similar conditions attract not just overseas investors but buyers with more explicit speculative motives (with short-term holding expectations simply to exploit price rises). Buyers from Egypt and North Africa became a significant impact on the top quarter of London’s market 2014-16. In the UK, Canada and Australia, Chinese investors have been regarded as key sources of speculation in, London, Vancouver and Melbourne, and especially in downtown locations. Policy action on this specific issue preceded any serious research in some of the cities we examined and this was somewhat unfortunate because the empirical evidence on ‘foreign buyers/speculators’ is quite mixed. The evidence on Canadian cities is that, since 2015-16, the rate of purchases by foreign buyers is usually less than 3 pc in metropolitan markets. However, there are some neighbourhoods-submarkets within the major metropolitan areas that in measurement periods displayed foreign purchase rates between 7 and 10 percent. Usually these neighbourhoods were within central cities and were in more expensive single family homes. There is little evidence of these properties being held vacant and they are either being used as a principle residence, to house children studying in Canadian universities, as a source of rental income or as a second home. The numbers involved in ownership simply for speculation appear to be a thin additional layer of demand in already pressured markets. Stripping away that layer will not resolve the pervasive affordability crisis as it seems to relate to fundamental market drivers reflecting domestic demands of different kinds. For instance, in Vancouver, net immigration from other Canadian provinces is quantitively a more significant addition to metropolitan housing demands. There is housing market evidence from the UK and the USA that suggests quite small numbers of additional searchers in particular submarkets of metropolitan housing markets can generate much wider price effects through expectational effects but this is most likely where housing markets are already tight.
- Price booms don’t last forever. In the Shaping Futures group, London and Edinburgh, after long periods of steady upward price shifts, have seen reduced house price growth as constitutional reform dominates British economic and political debate. In Australia price rises for Perth and Brisbane have remained above the CPI but are at lower rates than for the rest of the millennium. In Canada, as mentioned above, Calgary and Edmonton have experienced reduced inflation rates as oil prices fell and remained flat. In none of these markets have house prices spiralled downwards in the ways the IMF, and now national Finance Ministries, seem to expect (Vancouver prices are already rising again after the policy ‘shocks’ on 2016).
Debt and Instability
The Canadian evidence on rising household debt, that has also been evident in the UK and Australia, has induced the Bank of Canada to urge caution in lending growth (with the RBA and the Bank of England indicating similar concerns). Such caution is wise but it is important that consequent measures do have significant unintended consequences and policy measures are well designed to avoid restrictive policy impacts in unpressured markets.
Much reference is made to the role that housing played in the GFC, and no doubt rising prices in the boom exaggerated the upswing and static or falling prices extended recessionary effects. However, in major respects none of these three countries is currently replicating the US experience of the period 2000-2008 when grossly risky lending systems laid the foundations for worthless US mortgage backed securities to internationalise the effects of US housing finance policy weaknesses and undermine major financial institutions in multiple countries. In the GFC period, it was not housing market failures and collapses in the UK, Australia and, especially, Canada that triggered instability.
Four institutions set the parameters for financial stability policy and management in Canada, namely the Department of Finance (DOF), the Office of the Superintendent of Financial Institutions (OSFI), the Bank of Canada (BOC) and the Canadian Deposit Insurance Corporation (CDIC) and they all report to the Minister of Finance and they also Chair all they key national committees relating to financial stability. After 2010 the Canadian Housing and Mortgage Corporation (CMHC) played a major role in supporting credit expansion, essentially developing roles in creating credit derivatives, to deliver economic stability. However, since 2012 it’s loan insurance has been subject to critical scrutiny by the other major financial authorities. A significant number of smaller mortgage lending trusts participated in the post 2010 mortgage credit expansion so that their assets consists of mortgages insured by CMHC (the government) and whose deposits, or liabilities, are guaranteed by CDIC (the government). Mortgage market difficulties could therefore mean a double contingent liability materialising for the Government of Canada.
The downside of scrutiny and strategy processes is that the substantial housing market insights and competences of CMHC, separate from its insurance/portfolio management roles, lie outside the discussion of financial stability policy and housing market impacts. This seems an important and detrimental siloing of expertise within Federal government and this is reflected, perhaps, in a lack of housing economy knowledge within the supervisory authorities.
Central banks (and the IMF who issue frequent instability ‘alerts’) need to have regard to and knowledge of the ‘fundamentals’ of housing market dynamics as well as to their fears about financial stability and we have been unable to identify in the literature any econometric analysis of house price cycles in Canadian metropolitan areas over the last decade. There may well be a case, see further below, that the reality of the last five years of major metropolitan house price inflation and household debt increases have been fuelled not by loose, unconsidered mortgage lending, and pervasive speculation but by real drivers of housing demand in supply constrained locations. This does not mean rising loan to income or other measures of financial burden for households can be ignored but they need to be understood and their implications for periods when interest rates will rise considered.
The evidence of past research suggests that purchasers who live in their homes (unlike speculators) will go to extraordinary lengths to sustain their mortgage payments when rates rise but that shocks or policies that depress employment for mortgage holders are the most likely triggers to forced sales and attenuated prices. Even with these employment shocks, Canadian homebuyers have mortgage arrangements more akin to UK than US experience: buyers in difficulty cannot simply hand in the key and walk away to unwind mortgage debt. In many respects the IMF have not had a very accurate record in predicting bubbles and their unwinding in inflated markets. In the last five years, Oslo, Sydney, Vancouver and London have all attracted IMF comment. They all, like Toronto and Vancouver, have now slowed from price peaks in 2016 or earlier. None of these housing markets have collapsed and real demand is still driving market pressures (even in London the huge negative effects of the Brexit vote have slowed rather than collapsed price increases over the last year). How much understanding of real housing economics shapes these financial sector ‘warnings’ and decisions? Some commentators (Fraser Institute 2016, Benjamin Tal, 2016) have emphasised that it is important not to exaggerate financial instability concerns and we share that view. It is worthwhile to take a more detailed look at the growth in debt in Canada in recent years to reach a more nuanced view of housing market effects and prospects. Average gross household debt to income, as depicted in the figure 4 below, and indexed at 100 in 2000, rose sharply and steadily from 2002-2012, to around 150. The ratio has grown less rapidly since 2010 than before. Canadian growth has been broadly similar in level and pattern to Australia (indeed Australian figures exceed Canadian outcomes). It is important to note that for those households who had variable rate debts, debts they could refix or who had the deposit capacity to take new loans all benefitted from low interest policies post 2010. After then the ratio of debt servicing payments to disposable incomes, Figure 6 below, possibly a better measure of debt pressures on household budgets, has increased very little through the post 2010 period and the debt service ratio has remained relatively fixed. The most recent figures from Statscan (Cansim table 380-0073) provide a range of indicators of potential instability that suggest a somewhat less difficult position than the headline household debt to disposable income ratio of 168. When the assets held by households are taken into account the overall household debt to assets ratio fell from 16 to 14 pc through the 1990’s, rose to 18pc by 2010 and has since steadily fallen back to around its 1990 level of 16pc. The ratio of household debt to net worth fell by half a percent to 20pc through the period of peak house price increases from 2014-7(end quarter 2). These asset measures increasingly reflect housing wealth, with the real estate wealth to disposable income ratio rising from 405 to 427, so they cannot be assumed to be stable in the face of housing market instabilities. There has arguably been insufficient attention to how the distributional patterns of housing wealth, not least by age of owner, affect the security of households in the face of income and employment difficulties. Neither the IMF nor the Bank of Canada seem to have paid any regard to the increasing flows of housing wealth across different generations within families and these processes may help market stability (and sustain demand growth even when first home-buyer incomes fall behind house price growth). The roles of the Bank of Mum and Dad (BMD) (and Grandma and Grandpa) is part of the narrative of this debate in the UK and Australia with BMD (UK) contributing $10 billion in deposits for first time buyers in 2016. Canadian experience needs to be explored and the financial stability implications reviewed. A final note of concern regarding stricter, higher cost for mortgage holders, measures is that in the 2014-17 period non-mortgage debt payments by households grew faster than mortgage debt and were not primarily directed to long-term assets and had higher debt service ratios. The federal government seems to have aimed its stability concerns at housing when reducing non-asset creating debts, not least to support conspicuous consumption, would have served the nation better in the long term. Disrupting lifetime asset formation careers in housing rather than reducing excess demands for consumption or short-lived assets does not seem the optimal choice for future Canada.
|Figure 4. Household Debt-to-Income Ratios: Canada and Australia|
These patterns raise some doubts about the arguments of imminent collapse of ‘Canadian’ housing prices. This scepticism is reinforced when the ways in which different real metropolitan demand drivers have produced different real price and debt behaviours are considered. The empirical evidence is that real price rises have differed sharply across metropolitan areas but also in different submarkets within the fast inflating cities. Bank of Canada figures, displayed in Figure 7 above (BOC, 2016), indicate that the Canadian average debt to income ratio encompasses ratios below 125 in MB, QC, Atlantic and the North, and only at the end 2015 does it exceed 150 in BC. The pattern in Figure8, below, shows that in every major city in Canada the proportion of insured loans with loan to value ratios exceeding 4.5 has increased since 2013. In Calgary, Victoria, Vancouver and Toronto the proportion exceeded 20pc before 2013 and by 2015 was around 30pc in all these markets. However, in not one of the other locations had the 2015 level reached the 20pc of 2013 in the four pressured places. Other major cities and the rest of Canada were around 10pc by 2015. These spatial differences (which may have changed in the period after 2015 should be a paramount concern in national, provincial frameworks for housing policy choices. Governing central-local strategies for effective housing market policies is a key challenge for Canada.
3. Addressing Consequences
It is clear from the above section that sustained but different house price inflation rates, primarily driven by economic fundamentals, across metropolitan areas have had major effects on the Canadian housing system and the economy. Not all of these outcomes are supportive of major local and federal policy goals. There needs to be much more detailed research on what has driven metropolitan prices and different causes of supply inelasticities as well as the effects on housing affordability, inequality, commuting behaviours, more and longer renting by under-35’s, growing baby boomer ownership of rental units and a range of other housing outcomes that impact economic behaviour and productivity. Resolving these issues requires a range of significant reforms in Canadian housing policies and they are outlined in our concluding section. But for policy actions to date, however, it is financial stability/bubble formation concerns that have led to recent Canadian policy actions. If the national financial policy parameters are set and will have different impacts across cities and regions it is then essential to have a clear understanding of what housing market policies are required federally and provincially to achieve the housing outcomes to which they aspire.
4. Short Term Policies: Questions
The concern of Federal and Provincial governments to avoid housing bubbles in Vancouver and Toronto was well placed. However, there appears to have been a worrying absence of the use of empirical evidence, and especially of any economic or econometric analysis of national and regional housing markets, in policymaking to date. The recent report by the Finance Committee of the House of Commons (2017) addressing housing markets, lending and prices, is replete with different opinions about what has been changing in Canadian housing markets but at no point presents any evidence review that would inform public debate. Nor is it clear what empirical assessments, or evidence, of national and major metropolitan housing markets have justified recent changes in financial regulations and indeed tax and regulation changes for non-resident (foreign) house purchasers in Toronto and Vancouver (indeed in the Ontario case evidence appears to have been produced after the policy changes had already been made).
The foreign-buyer transaction tax changes implemented in Vancouver and Ontario (at different rates and at different times) appear to have cooled foreign buyer interest but strong price pressures remain. They represent, in a sense, emergency measures to deal with a poorly defined problem. We do not analyse them further and instead concentrate on the efficacy of the new nationally applied Federal mortgage rules. These policy changes, initiated to cool hot housing markets in Vancouver and Toronto, have rationed out many first-time home buyers in the previously well-balanced metropolitan markets in Canada. We have indicated above the extent to which relative balance prevails in much of metropolitan and regional Canada if not in the two major cities.
The measures required to cool booms driven by investor demands, and now associated with locally significant components of overseas investment, require coherent provincial and Federal fiscal policy responses for those booming areas, rather than ad hoc nationwide monetary policy/mortgage rules. National tools are inappropriate for inherently local markets. The October 2016 mortgage rule changes do not appear to have fundamentally slowed the housing markets in Vancouver and Toronto. However, they have undoubtedly created problems in metropolitan housing markets that were less obviously pressurized.
In Nova Scotia, for example, it appears that the flow of first-time home buyers into the market has been significantly reduced by the new deposit restrictions. CMHC has suggested that the new mortgage rules will take out 15-20% of the first-time buyer market in Atlantic Canada. A steady stream of households capable of sustaining first time entry into home-ownership is central to the functioning of the whole housing market system. Home purchases by such first-time buyers allows others to move upmarket and adjust their housing to changing household circumstances. Without first-time buyers, the system slows down. By moving out of rental units, first-time home buyers also create vacancies for those following in their footsteps into independent living. Putting an unnecessary brake on first-time home buyers in regional housing markets is likely to hamper their residential and economic progress.
CMHC describes the Nova Scotia housing market as “weak”. Unwittingly exacerbating a weak housing market has a negative effect on the regional economy: Decent housing at affordable prices is one of the ways in which less pressured areas of Canada can attract an increasing number of younger households who are relatively ‘footloose’ in their job location choices, and who want to have an affordable family life in home-ownership. With new ways of working online, moving to affordable, regional Canada is one important route for a subset of the workforce, for example, knowledge industry workers to move from pressured metropolitan areas to currently slower growing regional towns and cities. Federal monetary policy is, arguably, stopping this potentially important inter-regional adjustment mechanism, just when it would work most effectively. It may also be hampering, the ability of young Nova Scotian first-time home buyers from settling and committing to communities in the province. These are not just issues for Nova Scotia, nor just Atlantic Canada but the nationwide set of better balanced housing markets.
Delaying housing ownership for the young, where they have the incomes to sustain long term ownership, distorts household choices now and raises significant questions for future family life-courses. None of these considerations seem to have been considered in designing and understanding any modelling undertaken within government. The many participants in the House of Commons Committee Report did not raise key questions of housing economics, and we pose nine areas of such questioning for consideration by the Department of Finance, CMHC and the Bank of Canada:
- First, have they undertaken any metropolitan and regional market level modelling or analysis of the potential impact of the new mortgage rules and assess how they differ by region?
- How will the new rules delay or even ultimately deter the entry of young Canadians into home-ownership?
- What is the cost to asset accumulation, and family formation, for first-time home buyers by pushing such choices further back in their life-cycle?
- Will the rising proportion of Canadians paying mortgages into their 60’s and 70’s continues to increase, and what will this do to patterns of old age support and well-being?
- We would like to see consideration of the consequences of preventing first-time home buyers from getting into the market in periods of historically low interest rate environments, both in terms of the benefits of higher monthly loan service debt retirement in low interest rate settings, and intergenerational equity and in relation to the impacts of savings and consumption by these households in the decades ahead when they will be potentially making mortgage payments whilst other life cycle consumption demands may be rising.
- With regards to stated federal government concerns related to household indebtedness;
- We seek a much clearer explanation of the ways in which the Department of Finance and the Bank of Canada use the household debt to income ratio to assess housing ‘bubbles’ in a context where gross household debt figures are used to make contrasts with the past, but where first-time home buyer’s both have larger equity deposits, and have growing home equity as prices rise.
- It is too simplified to conclude that first-time home buyers will contribute to increasing high household indebtedness in anyway, let alone any meaningful way without better understanding how;
- household indebtedness differs among young people who own versus some other form of habitation,
- what their spending habits would be in the alternative of ownership,]
- how mortgage and non-mortgage debt has changed in the past number of years,
- whether household indebtedness manifests itself differently across regions and demographic groups, and
- if indeed there should be a greater concern for household indebtedness in Vancouver/Toronto than in non-pressured markets.
- We need a much more elaborate assessment, in the current functioning of housing markets, of how many likely defaults on mortgages there will be for every 1,000 first-time home buyers that enter the market. It is important to know the following.
- What is the government’s estimate of how many first-time home buyers the new mortgage rules have rationed out of the market since October 2016?
- In the absence of a mortgage, what are the interim alternative savings and spending habits of such prospective first-time home buyers?
- When is it expected that those rationed out of the current housing market, and who may have rates of return on savings below house price appreciation rates in their metropolitan markets, will be able to achieve ownership? This leads to a key housing policy question: will expanding market rental alternatives for this group, to meet their currently achievable housing demands, improve their wellbeing if house price inflation rates do exceed rates of return on savings (for then households will face an increasing hurdle to shift from renting to owning): we see no clarity on this critical policy issue in current debates.
- The House of Commons report indicated that the average mortgage default rate is 0.27 and rose to over 0.5 in the GFC. In aiming for some ‘safe’ borrowing pattern how do federal authorities weigh the benefits of maintaining, say, the 0 .5% national default rate on mortgages against the costs of rationing out significant proportions of first time buyers. Has the government modelled the relative benefits of getting 995 out of 1000 households safely into asset accumulation versus the costs of 5 in 1000 losing their homes? What informs this decision calculus?
- Have governments at federal and provincial levels seriously modelled the effect that moving into home ownership will have on their complimentary goal of freeing up more rental opportunities?
- What are the pros and cons of various branches of the Federal Government encouraging investment in appreciating assets such as homes, and in discouraging investment in depreciating assets such as cars?
- Has there been an attempt by Federal Government Departments to measure the historical impact that home ownership has had on productivity in regional economies and on individual asset accumulation. What do housing market outcomes do to productivity growth and inequalities of wealth within (and between) metropolitan and rural Canada?
- Has the government analysed the decrease in participation by first-time home buyers in the market since the changes to amortization length and down payment rules, namely 40 year amortizations to 25, and 5% deposits to 20%, and what specific delays that has had on their aspirational goals to own a home. It is important to understand the cumulative and granular effects of these changes.
5. What about the long term; addressing the Policy Fundamentals
The questions we have posed above indicate both the importance of an orderly housing system in which Canadians make choices, live their lives and accumulate assets and the significant knowledge gaps that now exist for sound knowledge-influenced policy making. Even more gaps exist if we look at longer term, more fundamental changes. In this brief discussion of the future we draw attention to policy questions that pertain to rebalancing the housing system so that it reflects, as far as possible, the preferences of households (and is therefore primarily, but not exclusively, market driven), becomes a mechanism for saving rather than speculation and facilitates Canadian productivity growth by being flexible as possible in the face of change. That is, the problems of Canadian housing markets, are not primarily about foreign buyers in boom metropolitan markets, but in the neglect of rising real house prices as a sign of both growing system pressures and supply side failures over decades. In many respects house price inflation has been as corrosive to productive growth as wage inflation was in the 60’s and 70’s but there has been an absence of will to tackle the issues.
There are some very general points of debate to move understanding of Canadian housing markets forward. Our ten key points for the long term include:
- Creating new alertness to the drivers and consequences of house price increases that outpace the general inflation rate
- Recognising that the critical locus for designing detailed housing strategies is at the metropolitan or regional housing market scale
- Having all orders of government acting together and with CMHC having a renewed role in facilitating these multi-order partnerships
- Placing housing investment policy within the framework of infrastructure policy, and seeing housing as essential economic, social and environmental investment
- With multi order involvement in housing investment, delivering combined levels of policy resources (money, staff capacity, land) via vehicle such as metropolitan-regional housing deals, may be the best way forward to align interests and reduce past behaviours that have in the past wasted housing policy resources, sometimes in non-housing purposes
- Seeking to minimize the burden of paying for housing policies on the productive activity of firms and households: this involves removing regulatory distortions that raise housing and land prices without apparent benefits but also recognizing that the vast unearned benefits of rising land and house prices accruing to existing owners are often simply ‘scarcity rents’ and that inclusionary zoning should be used to remove them when they are excessive
- The efficiency and fairness of taxation of housing need to be assessed for the longer term
- Where there is a policy need to facilitate the progress of poorer and younger households from renting into ownership action should be associated by supply side measures; demand side subsidies should simply be avoided
- There should be growing contestability of roles and subsidies between market and not-for-profit providers in areas of active policy to ensure that the long-sightedness and patient capital in non-profits and the acuities of market behavior might have the chance to cross-fertilise more innovative ways to address market failures as well as low income housing problems.
- Developing much-needed applied research capacity in Canadian housing, with data and evidence available in the public domain and, in that regard Canadian Housing market interests, governments and academics should look to the Australian AHURI model as a policy relevant and industry influenced model for delivering high quality housing research and debate in a large, federal country.
These are complex and contentious issues but we believe that Canadian housing policy debate has still not grasped some of the fundamental issues for a stable, efficient system for the long term. We want to move to a housing system consistent with a society rewarded by effort and entrepreneurship, and saving, rather than one where unearned gains divide rich and poor, young and old, city and countryside. We argue for more efforts in Canada to adequately identify problems and use appropriate market evidence in policy formulation, highlight weaknesses in recent short term policy and outline possible longer term measures to address the complex problems involved.
The Hon. Scott Morrison, Federal Treasurer of Australia recently noted, in a speech in Melbourne, that ‘home-ownership had been good for the country’ and a similar sentiment could be expressed for Canada. He also noted with concern that home ownership has fallen from 71% to 67% in the last two decades, and between 2002 and 2014 home ownership among 25-34 year olds declined more significantly by almost 10% points to less than 30%. That is more than 160,000 young people that would otherwise now be home owners. Many of these young people will have delayed their home ownership. Some may never get into the market. It is important to think carefully about the consequences of letting something similar happen in Canada.
It is also important to recognize that, whilst ownership growth may meet aspirations, wider economic questions surround the consequences of ownership expansion, and sustained rises in real house prices. We are not, emphatically, arguing for a fall in house prices and we are not of a view that a national housing ‘bust’ is likely in the foreseeable future. We are however, arguing for a much more coherent Canadian approach to housing market policies that sets house prices as a key area of policy interest at and between local, provincial and federal levels and that pays much more attention to researching, modelling, debating and governing price policies.
Bank of Canada, 2016. Annual Report. Ottawa
Chowdhury. R.A. and Maclennan. D. 2014 Regional House Price Cycles in the UK, 1978-2012: A Markov Switching VAR. Journal of European Real Estate Research, 7, 3, p345-366. 22p.
Di Matteo. L. 2017 Household Debt and Government Debt in Canada. Fraser Institute, Calgary.
Harcourt. M. 2007 From Restless Places to Resilient Communities. Report of the Prime Minister’s External Advisory Committee on Cities.
House of Commons. 2017 Canada’s Housing Markets: Benefits, Barriers and Bringing Balance. Report of the Standing Committee on Finance. Ottawa.
Hulchanski. D. 2015 Toronto’s Growing Socio-Spatial Divide: An Update on the ‘Three cities in Toronto’ Trend, 1970-2012.
Lowe. P. 2017 ( Governor of the Reserve Bank of Australia). Australia and Canada – shared experiences. Speech at the Australia-Canada Economic Leadership Forum, Sydney, 22 February 2017.
Maclennan. D and Miao. J, Housing and Capital in the 21st Century. 12th March, 2017. Housing, Theory and Society. Latest Articles. 19p.
Morrison, S. 2017. Speech to AHURI Conference, Melbourne 25th March, 2017.
Puga, D. 2010 The magnitude and causes of agglomeration economies. Journal of Regional Science, 50(1), 203-219.
Tal, B 2017 On the Fallacy of the Debt-to-Income Ratio.The Week Ahead.March 20-24, 2017
CIBC Capital Markets, Toronto
Glaeser. E. (2008) The Economic Appraoch to Cities. Arvard Institute of economic Research discussion Paper No. 2149
 David Graham is the President of Atlantic Developments (Halifax, Nova Scotia). Duncan Maclennan is Professor of Public Policy at the University of Glasgow and Professor of Strategic Urban Management and Finance at the University of St Andrews, Scotland.