At 103% of GDP, Canadian households have the most debt in G7

Ask The Hub

Ask The Hub The article suggests Canadian debt is unproductive. How does it define 'unproductive' in this context, and what are the consequences? What policy changes does the author suggest could help address Canada's high household debt and its impact on the economy? The article compares Canada's residential investment to other countries. What are the key differences, and why are they significant?

In its latest Annual Risk Outlook, released on April 14, the Office of the Superintendent of Financial Institutions (OSFI), Canada’s banking regulator, identifies real estate secured lending and mortgage risk as the top threat facing the financial system. The regulator expects residential mortgage arrears and defaults to rise over the next two years. This warning underscores a question that deserves more sustained attention. Canada’s household debt is perennially in the spotlight, and for good reason. At 103 percent of GDP, Canadian households now carry the largest debt burden among G7 nations, and the second-largest among advanced countries after Switzerland. Total household credit market debt reached $3.2 trillion by the end of 2025, or 177 percent of household disposable income. For every dollar of disposable income, Canadians owe nearly $1.77 in credit market debt. These figures are alarming on their own. But the standard defence goes something like this: most household debt is mortgage debt, and mortgages finance an appreciating asset. Unlike credit card debt or consumer loans, borrowing to buy a home builds wealth over time. Roughly 75 percent of Canadian household debt is mortgage debt, so the argument goes that the headline numbers overstate the risk. Canadians are leveraged, yes, but they are leveraged against something of enduring value. This defence has significant limitations, and the policy conversation it supports has been stuck on a familiar question. The debate centres on whether household debt levels are sustainable and whether a correction might trigger a financial crisis. Those are important questions. But a more fundamental question is whether the debt is productive. When most of a nation’s household debt is concentrated in a single asset class, the risks are not diversified away. And when the borrowed capital flows predominantly into existing housing rather than productive investment, the economic cost extends well beyond household balance sheets. The defence fails on its own terms Consider the logic of this standard defence at scale. Any individual household can, in theory, sell a home to service its debts. But all Canadian households cannot do this simultaneously. If a large number of homeowners tried to liquidate at the same time, prices would fall, and the debt obligations would remain fixed. At the national level, the aggregate housing wealth that backstops $2.4 trillion in mortgage debt is ultimately supported by other households’ willingness to take on similar levels of debt. The reasoning is circular. Leveraged wealth amplifies losses in ways that unleveraged wealth does not. A household that purchased a home with 20 percent down at the peak in 2022 and subsequently sees a 20 percent price decline has lost the entirety of its equity position, while the full mortgage balance persists. Housing wealth built through borrowing is fundamentally more fragile than wealth accumulated through savings or investment in financial assets. The 2022–23 interest rate hiking cycle provided a preview of this fragility. The household debt service ratio peaked at 15.3 percent of disposable income in 2023, and the effects continue to ripple through the system. According to the OSFI report, 3.1 million mortgages, or 52 percent of the total, will renew by the end of 2027. Of these, 1.3 million are fixed-rate or variable-rate fixed-payment mortgages that will be renewing for the first time since the low-rate period of 2021 and 2022. OSFI expects these borrowers to face material payment increases. The Canada Mortgage and Housing Corporation (CMHC) reports that 1.5 million households have already renewed at a higher rate in 2025, and mortgage arrears are rising across multiple segments, particularly in Toronto and Vancouver, though arrears remain below historical peaks. The condo segment is especially strained, with OSFI noting that sales in Toronto and Vancouver have fallen to levels not seen since the 1990s, and many new units are now worth less than their presale purchase prices. The fragility extends across generations. A recent Bank of Canada analysis found that parental co-signing of first-time buyer mortgages has nearly tripled since 2004, with three-quarters of co-signed borrowers unable to qualify on their own, spreading mortgage risk from one household balance sheet to two. To be fair, most borrowers appear to be managing. It has not been a cliff in the catastrophic sense. But even an orderly adjustment redirects household income toward debt service and away from consumption, savings, and productive investment. The leverage constrains flexibility even when it does not trigger default, and that constraint compounds across millions of households at the same time. What the debt actually finances The deeper problem with the mortgage defence becomes apparent when examining what all this borrowing actually produces. Canada’s national accounts classify “residential investment” as comprising three components: new construction, renovations, and ownership transfer costs. That last category includes realtor commissions, legal fees, and land transfer taxes. When a household purchases an existing home, the transaction itself does not appear in GDP since it is merely an asset transfer. But the associated service fees do, and they are recorded as “investment.” This matters because it means a significant share of what Statistics Canada labels residential investment creates no new productive capital. It represents transactional churn: the cost of shuffling ownership of existing housing stock. The realtors, lawyers, and land transfer offices perform real services, to be sure, but classifying these fees as investment obscures a critical economic distinction. That distinction matters because it shapes how we understand what is happening to Canada’s investment mix. The scale of this reorientation toward housing is visible in nominal GDP shares. In 2000, residential structures accounted for 4.3 percent of GDP, while investment in machinery, equipment, and intellectual property collectively stood at 8.3 percent. By 2024, those positions had effectively reversed: residential structures had risen to 7.6 percent of GDP while machinery, equipment, and intellectual property fell to 5.7 percent. Stripping out intellectual property investment reveals an even sharper decline: machinery and equipment alone almost halved from 6.3 percent to 3.3 percent of GDP, as residential structures nearly doubled. Much of the increase in residential “investment” reflects the dramatic inflation in housing prices over this period, but the result is the same: a growing share of Canada’s nominal income now flows through housing rather than the productive capital that drives long-term growth. Of the residential investment that does occur, only about half represents new construction. Roughly a third flows to renovations of existing properties, and about a fifth covers ownership transfer costs—realtor commissions, legal fees, and land transfer taxes associated with resale transactions that create no new housing stock. To place this in an international context, Canada’s residential investment peaked at nearly 10 percent of GDP in 2021. The United States was widely considered dangerously concentrated in housing at about 6.5 percent of GDP in 2006, before the crash that triggered the global financial crisis. The two countries have materially different lending rules, but the scale of the gap is telling. Among Organisation for Economic Co-operation and Development (OECD) countries, Canada dedicated the largest share of GDP to residential investment over the 2018 to 2023 period, averaging 8.3 percent. That is nearly double the United States (4.2 percent), and well above comparators like Australia (5.4 percent), the Netherlands (5.4 percent), and the United Kingdom (4.0 percent). The domestic investment composition data tells a similar story. Between 2014 and 2021, housing comprised 34.1 percent of total investment in Canada, compared to 18.5 percent in the United States. Over the same period, investments in information and communications technology accounted for 10.4 percent of total Canadian investment versus 16.5 percent south of the border. These are not marginal differences. Housing is not the only sector that has expanded at the expense of productive business investment. Former Bank of Canada Governor Stephen Poloz recently argued on the WONK podcast that federal government spending, which rose by roughly 3 percentage points of GDP from the last Harper budget onward, represents a significant source of crowding out in its own right. In his telling, the expansion of the public sector displaced employment and investment in more productive sectors like resources and manufacturing, compounding the drag from the energy price collapse that began after 2014. The full picture is likely more complicated than any single explanation allows, but the direction is consistent: whether through housing, government expansion, or some combination, capital and labour in Canada have been migrating away from the sectors that generate productivity growth. The productivity cost The consequences of this capital allocation pattern are visible in productivity data. Labour productivity in residential construction fell 37.3 percent cumulatively from 2001 to 2023, declining at an average rate of 2.1 percent per year. Over the same period, the overall business sector saw productivity grow by 12.5 percent or 0.5 percent each year. The sector absorbing a disproportionate share of Canada’s capital and credit is becoming less efficient at producing output. The mechanism connecting household mortgage debt to weak business investment appears to operate through the financial system. Analysis of financial flow data by Alberta Central suggests that from the mid-1990s onward, the household sector absorbed most of the net lending available in the Canadian economy, primarily through mortgage borrowing. The corporate sector, by contrast, has been effectively squeezed. In earlier decades, government deficits absorbed available capital during the high-deficit era of the 1980s and 1990s. Since the mid-1990s, household mortgage debt appears to have played a similar role, contributing to chronic underinvestment in machinery, equipment, and intellectual property. McKinsey’s analysis of Canadian productivity confirms the broader trajectory: productivity growth has shifted from export-oriented sectors toward real estate and domestic industries. Canada has, gradually and perhaps inadvertently, structured a significant portion of its economy around a sector that contributes relatively little to long-run prosperity. Reframing the question The evidence reviewed here suggests the trillions of dollars Canadians have borrowed and channelled predominantly into housing are not generating the kind of economic returns that justify the associated risks and opportunity costs. Several features of Canada’s policy environment actively encourage this pattern. The unlimited principal residence capital gains exemption makes housing the most tax-advantaged asset class in the country. While most advanced nations also exempt primary residence gains, the OECD has recommended capping these exemptions to reduce the incentive to overinvest in owner-occupied property. Risk-weight differentials in financial regulation make mortgage lending more profitable for banks than commercial lending, structurally tilting the flow of credit toward housing. OSFI Superintendent Peter Routledge has acknowledged the distortion, noting that banks’ internal risk weights for uninsured residential mortgages run around 10 to 12 percent, compared to 50 to 60 percent for corporate and small business loans, meaning a bank needs roughly five dollars of capital to back a business loan for every one dollar behind a mortgage. And CMHC’s mortgage insurance framework socializes downside risk while leaving gains with borrowers and lenders, reinforcing the incentive to lever up. These are policy choices, and difficult ones at that. Until policymakers grapple seriously with the composition of household debt, with what it actually finances and what it crowds out, we will continue treating symptoms while the underlying condition persists. The mortgage myth offers little comfort.