HOUSING OBSERVER. An Examination of Household Indebtedness. Article 2 March 2016

Transcription

1 HOUSING OBSERVER 2016 Article 2 March 2016

2 Table of Contents 1 Overview of Canadians financial health Changes in household borrowing Looking ahead: implications of the changing composition of household balance sheets Appendix...13 Although this information product reflects housing experts current knowledge, it is provided for general information purposes only. Any reliance or action taken based on the information, materials and techniques described are the responsibility of the user. Readers are advised to consult appropriate professional resources to determine what is safe and suitable in their particular case. Canada Mortgage and Housing Corporation assumes no responsibility for any consequence arising from use of the information, materials and techniques described. 2

3 Over the past decade, Canadian households have become increasingly indebted, with total household credit edging higher again in the third quarter of 2015 to $1.9 trillion. 1 Mortgage credit, at over 70% of total debt, was the principal driver of this increase. Between 2001 and 2011, average inflation-adjusted MLS residential housing prices appreciated by 72%. 2 Without a corresponding rise in household income, the need for larger mortgage loans to finance housing purchases has increased. 3 As of September 2015, mortgage credit outstanding stood at $1.3 trillion, a year-over-year increase of 5.9%. This is below the average annual growth rate of 7.6% for , but higher than the 5.0% year-over-year growth in the previous year. Changes introduced by the Government of Canada to mortgage insurance rules aimed at tightening lending standards have contributed to the moderation in mortgage growth. 4 However, with the recent rounds of interest rate cuts, households borrowing behaviour continues to come under heightened scrutiny and remains an area of interest to policy makers and the general public alike. FIGURE 1 Household Debt to to income, debt to asset ratio and debt to net worth ratios, 1999Q Q3 Debt as a % of disposable income Debt to disposable income Debt to total assets Debt to net worth Debt as a % of assets Q Q Q Q Q Q Q Q Q Q Q Q Source: Statistics Canada (CANSIM Table ) 1 StatsCan (CANSIM table ). 2 MLS is a trademark of the Canadian Real Estate Association. 3 Median after-tax household income grew by 11% over the same period. Source: CMHC, adapted from Statistics Canada (Survey of Labour Income and Dynamics). 4 For an overview of the list of changes regarding government-backed mortgage insurance, see the Housing Finance chapter in the 2014 Canadian Housing Observer. 3

4 1 Overview of Canadians financial health The financial health of Canadians has received widespread attention in the media as household debt reached levels that many commentators fear are unsustainable in the long run. Since 2008, the Bank of Canada has listed the balance sheet of the household sector as a source of vulnerability, noting in the June 2015 Financial System Review that this vulnerability is edging higher due to an increase in the level of household debt and the negative effect on incomes resulting from the sharp decline in oil prices. 5 Household debt load has risen but remains manageable for most Canadians While debt holdings have undeniably risen in the past decade, several indicators suggest that under current conditions, the financial prospects for most Canadians remain healthy. Much attention has been paid to the aggregate household credit market debt-to-income ratio (Figure 1), which stood at a record high of 163.7% in the third quarter of 2015, compared to 106% in In other words, Canadians owed $1.64 for every dollar of annual disposable income. The ratio grew rapidly during the 2000s due to historically low interest rates and increased access to household credit, while the economic environment kept income growth muted. More recently, debt growth has moderated an encouraging sign that overall, Canadians are making efforts to better manage their balance sheets. The Spring 2015 Financial Industry Research Monitor (FIRM) Residential Mortgage Survey reported that about one-third of borrowers have made prepayments towards their mortgage principal, compared to one-fourth in the previous year. 6 Steady increases in asset values have generally kept debt-to-asset and debt-tonet worth ratios stable. However, the 2008 financial crisis resulted in marked deterioration in both measures. Debt-to-net worth increased from an average of 19.7% in 2006 to 23.4% in It has since improved as asset values once again returned to an upward trajectory. In Q3 2015, the debt-to-net worth ratio was 20.5%, marginally above the average of 20.1%. 5 Financial System Review. Ottawa: Bank of Canada, June uploads/2015/06/fsr-june2015.pdf (July 3, 2015). 6 The Financial Industry Research Monitor (FIRM) Residential Mortgage Survey. Toronto: prepared for CMHC by Altus Group Consulting and Ipsos Reid, Spring 2014 &

5 Understanding the Limitations of Debt Ratios The debt-to-income ratio compares the flow of household aggregate annual disposable income against a stock of debt that is usually repaid over several years. Since income is much more responsive to changing economic conditions than the stock of debt, the ratio can be highly pro-cyclical, deteriorating during an economic downturn. Further, the debt-to-income ratio does not take into consideration that debt may be accumulated to increase household wealth through the acquisition of assets. Although a ratio of 164% may appear daunting, it provides very limited indication of the extent of individual households' financial stress. The debt-to-assets and debt-to-net worth ratios are measures of financial leverage that show how well debt is secured by assets or equity. Neither ratio takes into account the liquidity of the assets i.e. the ease and speed at which assets can be converted into cash without significantly affecting price. Further, both ratios are highly dependent on asset prices and may understate the extent of households exposure to excessive debt during a period of expansionary asset prices. The ratios can deteriorate quickly when asset price falls, as witnessed during the recent financial crisis. The debt-service ratio underscores Canadians ability to service their debt. It is the percentage of annual disposable income that must be allocated toward annual interest payments on existing debt. Some estimates of debt-service ratios also capture principal repayments. Debt-service ratios are affected by movements in interest rates. Thus, ratios today should not be considered in isolation in assessing how well households will meet debt obligations in the future. This ratio in part underscores the significant equity Canadians have in their homes, averaging over 70% of the value of the home. 7 A higher level of home equity improves a household s ability to absorb fluctuations in house prices. Assuming a 20% and 40% drop in the market value of properties, DBRS finds that Canadians home equity would decrease to 63% and 51% respectively, still higher than the home equity ratio in the U.S. during the financial crisis. 8 Additionally, mortgages in Canada, unlike in the U.S., are typically full-recourse loans, which means that the borrower is responsible for repaying the loan even in the case of foreclosure. 9 This feature is expected to reduce the risk of households defaulting on their mortgages. Despite rising debt levels, low interest rates have kept the debt-servicing cost manageable. In Q3 2015, the mortgage debt-service ratio () i.e. the ratio of annual mortgage debt interest costs to annual disposable income stood at 3.28%, a substantial decline compared to 4.35% in 1999 and lower than the average of 3.97% (Figure 2). The total has also been in a declining trend since 2008 and at 6.33%, was more than a percentage point below the average of 7.58% and well below the 2007 high of 9.12%. 7 Dominion Bond Rating Service (DBRS) Canadian Housing Indicators. Toronto: DBRS, June Ibid. 9 A full-recourse loan allows a lender to seize other assets that belong to the borrower in the case of a default and the proceeds from the sale of the house are not enough to cover the outstanding loan balance. 5

6 Per cent FIGURE 2 Household debt-service ratios (interest only), 1999Q Q3 Total Mortgage Non-mortgage Q Q Q Q Q Q Q Q Q Q Q Q Note: Dashed lines are averages from 1999 Q1 to 2015 Q3 Source: Statistics Canada (CANSIM) To better reflect the actual burden of debt, Statistics Canada recently released new estimates of the s that account for both the interest and principal repayments. In recent years, these ratios have remained fairly stable but have not fallen in the low interest rate environment (Figure 3). The mortgage (interest + principal) was 6.15% in Q3 2015, unchanged from the previous quarter but higher than the average (5.71%). The divergence in the two measures of s suggests that Canadians are taking out larger loans thus increasing the amount of principal repayments required and have shortened amortization periods. FIGURE 3 Quarterly household debt-service ratio (interest + principal), 1999Q Q3 Per cent Total Non-mortgage Mortgage Q Q Q Q Q Q Q Q Q Q Q Q Note: Dashed lines are averages from 1999Q1 to 2015Q3 Source: Statistics Canada (CANSIM) 6

7 Proportion of financially vulnerable households is on the rise While debt holdings are higher on average, the degree of this increase was not experienced proportionately across the population. In 2014, about 12% of households had a total debt-to-income ratio above 250%, double the proportion in These households carried nearly 40% of total household debt. In particular, households headed by someone under the age of 35 made up around 30% of households with debt-to-income above 250% and held about 30% of aggregate household debt. 2 Changes in household borrowing Aggregate measures such as the debt-to-income and debt-service ratios provide a good gauge of the overall state of Canada s financial landscape; however, they do not address the underlying distribution of debt and do not provide information on the households most vulnerable to economic shocks. The remainder of the article will use data from the 1999 and 2012 Survey of Financial Security (SFS) 11 to analyze the changing patterns of debt holdings among Canadians in different age groups. Older households are more likely to carry debt today... The 2012 SFS found that just under one third of Canadian households held no form of debt. Although the percentage of households with debt has remained stable at the lower end of the age bracket, people are increasingly carrying debts as they age (Figure 4). In 2012, 72% of households aged 55 to 64 had debt of some kind, up from 61% in More strikingly, while less than a third of households aged 65 and over carried debt in 1999, this number jumped to 43% in Overall, households aged 55 and over held more than 20% of total debt, compared to 12% in 1999; however, this increase is in part a consequence of underlying demographic trends. 12 More importantly, debt burden for older households have increased. In 2012, median debt for households in the 55 to 64 group was $50,000, compared to $26,200 in Higher debt in the pre-retirement group has direct implications for the likelihood of carrying debt during retirement. For households aged 65 and over, median debt load increased 112% to $18,000 in This growth in debt comes 10 Financial System Review. Ottawa: Bank of Canada, December The Survey of Financial Security is an occasional survey conducted by Statistics Canada that collects information from a sample of Canadian families on their assets, debts, employment, income and education. The survey has been conducted in 1999, 2005 and Between 1999 and 2012, the number of Canadians over the age of 55 grew by more than 50%. A counterfactual distribution based on data from the Canadian Financial Monitor suggests that the share of total debt held by older households between 1999 and 2010 would have remained relatively unchanged in the absence of population aging. However, since total debt is higher, today s seniors remain more indebted than their counterparts a decade ago. See Allan Crawford and Umar Faruqui, What Explains Trends in Household Debt in Canada? Ottawa: Bank of Canada,

8 at an age when individuals usually experience a reduction in income, raising questions about their ability to service these debts while meeting other consumption needs. Previous research has found that retirees with higher annual income are more likely to have debt than retirees with income less than $25, The prevalence of elevated debt among high income seniors is of lesser concern as greater income flows will provide constant support to meet debt obligations. FIGURE 4 Household Percentage debt-service of households ratios with (interest debt by only), age, 1999Q1 and Q Per cent Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) Comparing Median and Average Values of Debt and Assets Where applicable, this report presents median debt and asset values, which are a better representation of the financial position of a typical household. The SFS data display significant discrepancy between median and average values. For instance, while median debt for households aged 65 and over was $18,000 in 2012, the average debt was considerably higher at $61,700. This discrepancy suggests that some older households are more at risk of elevated debt. 1 The difference between median and average values is also observed for other age groups (see Appendix). 1 Using data from the 2009 Canadian Financial Capability Survey, Marshall (2011) found a similar result for retirees 55 and over. Of retirees with debt, one-quarter owed less than $5,000, one-third owed between $5,000 and $24,999, one-quarter owed between $25,000 and $99,999, and the remaining 17% carried over $100,000 of debt. 13 Katherine Marshall, Retiring with Debt. Ottawa: Statistics Canada,

9 ... and are more likely to hold lines of credit While the distribution of total debt across different age brackets remained relatively unchanged between 1999 and 2012, largely reflecting underlying patterns in mortgage holdings, older households displayed an increased reliance on certain forms of debt. In particular, home equity lines of credit (HELOC) have grown in importance. In 2012, 8% of households aged 65 and over had HELOC debts, compared to less than 2% in The median outstanding balance of HELOCs was $36,000 in 2012, up from $26,200 in Overall, HELOC for this group represented 22% of total debt in 2012 (Figure 5). FIGURE 5 Breakdown of total debt, 1999 and 2012 Per cent Mortgages 2012 Mortgages 1999 HELOC 2012 HELOC 1999 Other LOC 2012 Other LOC 1999 Other Debts 2012 Other Debts Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) In contrast, the percentage of senior households holding mortgage debt increased modestly from 8% to 12%, with a median value of $80,000 in 2012 (up from $59,000 in 1999). Looking across age groups, the decline in mortgage debt as a percentage of total debt indicates that households continued to pay down their mortgage as they age (Figure 6). However, the increase in house prices in recent years likely bolstered the popularity of using home equity as a source of financing Results from the CFM similarly show a shift towards home-equity extraction, with 40% of the borrowed funds used to finance consumption and home renovation from 1999 to See Jeannine Bailliu, Katsiaryna Kartashova and Cesaire Meh, Household Borrowing and Spending in Canada. Ottawa: Bank of Canada,

10 Younger households are taking on larger debts due to bigger mortgage loans Debt holdings among younger Canadian households those between ages 15 to 34 and 35 to 44 have also increased, particularly for the latter group. The proportion of households aged 35 to 44 with debt edged up modestly (from 81% in 1999 to 87% in 2012), in stark contrast to the spike in the value of their debt. Median debt more than doubled, from $65,500 in 1999 to $146,300 in Mortgage debt dominated the debt portfolio of younger households as many are entering the housing market for the first time. The Spring 2015 survey by the Canadian Association of Accredited Mortgage Professionals (CAAMP) 15 reported that more than half of first-time buyers (57%) were between the ages 25 and 34 and another 18% were between the ages 35 and 44. Further, 62% of first-time buyers had down payments that were 20% or less. Consistent with rising housing prices and the consequent need to take out larger mortgage loans, median mortgage debt for households aged 35 to 44 was $190,000 in 2012, up from $91,700 in For households aged 15 to 44, mortgage debt was responsible for over 80% of the growth in total debt. In comparison, mortgages contributed to a little over half of the change in total debt for households aged 65 and above. Per cent FIGURE 6 Percentage of of households with mortgage debt by age group, 1999 and Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) 15 A Profile of Home Buying in Canada. Toronto: Canadian Association of Accredited Mortgage Professionals (CAAMP), June

11 Table 1: Composition of changes in debt and assets across age groups, 1999 and 2012 (millions of 2012 constant dollars) Changes Change in total debt ($) 145, , , ,164 59,620 Percentage due to (%) Mortgages Lines of credit Other debts Change in total assets ($) 361, ,579 1,172,411 1,594,589 1,281,709 Percentage due to (%) Financial assets Principal residence Other real estates Other assets Other debts include credit card and installment debt, student loans, vehicle loans and all other debts. 2 Financial assets include private pension assets and non-pension financial assets. 3 Other assets include other non-financial assets (e.g. vehicles, valuables and collectibles, copyrights and patents, etc.) and equity in business. Source: CMHC, adapted from Statistics Canada (Survey of Financial Security). Since debt was mainly deployed to finance purchases of real estate, the asset position of younger households improved. Median assets for households between the ages of 35 and 44 increased 77%, from $218,800 in 1999 to $388,000 in Real estate contributed to 66% of the rise in the total value of assets for households aged 15 to 34, and 87% for households aged 35 to 44. In comparison, for older households, real estate was responsible for less than half of the rise in asset values (39% for 55 to 64, and 44% for 65 and over) Financial assets (including private pensions) played a larger role in the overall valuation of assets for older families. 11

12 3 Looking ahead: implications of the changing composition of household balance sheets Aggregate data such as the debt-to-income ratio point to higher household debt in recent years. However, with 12% of households accountable for 40% of total debt in 2014, the proportion of highly leveraged Canadians remains concentrated. Low interest rates in recent years have made the cost of servicing debt largely manageable despite rising indebtedness. When the protection of low rates disappears, only a small group of Canadian households may find themselves in financial stress. Due to elevated mortgage debt and the significance of real estate assets in net worth, younger households are more susceptible to the risks associated with future interest rate hikes and house price fluctuations. One concern is that a sharp downturn in house prices can greatly erode the housing equity for this group, and when accompanied by a rise in unemployment or a drop in income, can trigger possible defaults and further ramifications in house prices. Nevertheless, with the fixed-rate, 5-year mortgage being the most popular among Canadian homeowners, the effect of an interest rate hike on debt-servicing burden will not be immediately felt for the majority of mortgagees. However, homeowners with a variable rate mortgage (about 30% of all outstanding mortgages) could be vulnerable. 17,18 Recent changes in mortgage rules will further reinforce the quality of mortgage credit and improve the capacity of Canadian households to deal with adverse economic conditions. For older households, the higher likelihood of carrying debt into retirement deserves attention. Faced with reduced income and an increasing debt load, older households may increasingly need to tap into the equity they have built up in their home. In this context, the share of HELOCs in total debt is likely to rise further in the coming years. A fall in house prices that leads to a loss of equity may require older households to adjust their consumption pattern. 17 Includes both variable rate mortgage and combination mortgage where a portion of the mortgage term or loan amount is at a fixed rate and the remaining portion is at a variable rate. Source: The Financial Industry Research Monitor (FIRM) Residential Mortgage Survey. Op cit. 18 It should be noted that homeowners have the option to switch to a fixed-rate mortgage when interest rates rise, though likely subject to fees and other costs. 12

13 4 Appendix Table A.1 Selected median and average debt and assets, households with debt or assets, 1999 and 2012 (constant 2012 dollars) All 15 to to to to and over Median ($) Total debt 40,000 66,900 31,700 50,100 65, ,300 51,400 91,000 26,200 50,000 9,200 18,000 Mortgages 90, , , ,000 91, ,000 78, ,000 65, ,000 59,000 80,000 Principal residence 87, , , ,000 91, ,900 76, ,000 63, ,000 52,400 67,000 Lines of credit 6,600 15,000 5,200 10,000 6,600 15,000 9,200 16,000 10,500 19,000 6,600 17,000 Total assets 227, ,900 65,500 85, , , , , , , , ,300 Financial assets 52,900 85,000 11,100 14,500 42,700 60, , , , , , ,000 Non-financial assets 142, ,500 37,300 39, , , , , , , , ,000 Principal residence 163, , , , , , , , , , , ,000 Net worth 155, ,100 36,400 42, , , , , , , , ,800 Average ($) Total debt 74, ,700 70, ,400 85, ,600 84, ,300 61, ,700 32,000 61,700 Mortgages 108, , , , , , , ,800 96, ,900 77, ,100 Principal residence 99, , , , , ,900 92, ,100 79, ,900 65, ,700 Line of credit 17,800 40,200 11,400 22,700 15,000 39,100 24,300 48,000 21,100 44,700 20,900 48,800 Total assets 387, , , , , , , , , , , ,900 Financial assets 170, ,300 40,400 63,800 94, , , , , , , ,500 Non-financial assets6 187, , , , , , , , , , , ,500 Principal residence 195, , , , , , , , , , , ,100 Net worth 336, , , , , , , , , , , ,300 Source: CMHC, adapted from Statistics Canada (Survey of Financial Security). 13

14 The Housing Observer Online offers a variety of information on different topics: Housing Markets Housing Finance Housing Data Housing Needs Housing Demographic Affordable Housing Seniors Housing Housing Policy Visit to get the latest information CMHC Project Manager: Xuetong Zhai, Policy Analyst, Housing Finance HOUSING MARKET INFORMATION PORTAL The housing data you want, the way you want it and when you want it... and it s all just a few clicks away. Free - Flexible - Comprehensive Visit cmhc.ca/hmiportal today!

15 Alternative text and data for figures Figure 1: Household debt-to-income, debt-to-asset ratio and debt-to-net worth ratios, 1999Q1-2015Q3 Quarter and Year Credit market debt to disposable income (%) Debt to total assets Debt to net worth Quarter and Year Credit market debt to disposable income (%) Debt to total assets Debt to net worth Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Source: Statistics Canada (CANSIM Table ) A1

16 Figure 2: Household debt-service ratios (interest only), 1 Canada, 1999Q1-2015Q3 Quarter and Year Total Mortgage Non-mortgage Quarter and Year Total Mortgage Non-mortgage Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Average (since 1999) Seasonally adjusted at annual rates. Source: Statistics Canada (CANSIM) A2

17 Figure 3: Household debt-service ratios (interest + principal), 1 Canada, 1999Q1-2015Q3 Quarter and Year Total Mortgage Non-mortgage Quarter and Year Total Mortgage Non-mortgage Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Average (since 1999) Seasonally adjusted at annual rates. Source: Statistics Canada (CANSIM) A3

18 Figure 4: Percentage of households with debt by age, Canada, 1999 and 2012 Age Year 1999 Year 2012 All Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) Figure 5: Breakdown of total debt, Canada, 1999 and 2012 Age Mortgages HELOC Other LOC Other debts Mortgages HELOC Other LOC Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) Other debts Figure 6: Percentage of households with mortgage debt by age group, Canada, 1999 and 2012 Age Year 1999 Year Source: CMHC, adapted from Statistics Canada (Survey of Financial Security) A4