Almost every working Canadian eventually wants to buy a home. And most end up doing so, given the home ownership rate is more than 66 per cent. Take out young people who can’t yet muster the down payment and the lowest quartile of income earners, and the ownership rate rises to about 90 per cent. It might come as a surprise, then, that owning is not always better than renting. Even more counterintuitive, renting would have been the better option for anyone who retired in the past decade, in spite of the hot housing market.
This conclusion is based on comparing a) buying a home and holding it for 30 years, with b) renting and investing an equivalent amount in the stock market. The analysis encompassed all 30-year periods from 1938-1967 up to 1990-2019.
In all cases, the household would have had total annual earnings of $110,000 in 2020. To determine historical pay levels, I deflated this figure by the change in the national average wage.
At the start of each 30-year period, the household is assumed to buy the most home they can with mortgage payments equal to 22 per cent of earnings in the first year plus a down payment equal to $80,000 in today’s dollars but deflated by the same factor as pay. For instance, a home buyer in 1990 would have bought a house worth $147,000 with a down payment of $38,000.
The renter is assumed to invest an amount equal to the down payment plus mortgage payments less the net rent for an equivalent home. Income tax paid by the renter on this side fund was assumed to be paid annually. To make it an apples-to-apples comparison, allowance was made for the property taxes, insurance and maintenance that a homeowner would pay explicitly. I assumed that the ratio of annual rents to house prices has fallen steadily, from 7 per cent of the current home price in 1990 down to a current 4.5 per cent. This ratio turns out to be a critical factor later when we analyze results.
As the chart shows, homeowners would have enjoyed a much higher net worth than renters in almost all periods ending before 2001. This reflects the fact that renters would have to dip into their savings to pay rent in the later years of each 30-year period as rents grow faster than mortgage payments.
The homeowner advantage shrinks dramatically for periods ending after 2002. After 2009, one would have been better off being a lifetime renter. To be clear, this statement refers to a young household making the decision to rent or buy between 1980 and 1990. The high mortgage rates during this period were a major factor that favoured renters.
As a caveat, the housing price index used in the analysis represents a composite of Canadian cities. Homeowners in Toronto or Vancouver would have fared somewhat better. For instance, the index indicates house prices have risen by 4.2 per cent a year in real terms from 2001 to 2020 whereas The Economist reports Toronto prices rose by an average of 4.6 per cent a year during the same period. On the other hand, the bubble-like conditions that have existed in Toronto and Vancouver for the past decade cannot last forever. House prices in these cities may remain perennially high but future increases in those prices will eventually subside as has happened in such places as New York.
This leaves us with the burning question of what should prospective home buyers be doing now? Is it better to rent or buy today if one has a 30-year investment horizon? To answer this, we have to look beyond the pandemic and consider the long-term prospects for stock market returns, house price increases and mortgage rates. Stock price valuations are currently very high, which may portend a prolonged slowdown in the not-too-distant future. One reason for the current high valuations is that interest rates are abnormally low, leaving nowhere else to invest. Interest rates are now stabilizing and will eventually inch up, which would be mildly negative for stocks. Increases in house prices should also slow down, though when that happens will depend largely on the strength of demand from foreign buyers. Also important is the ratio of annual rent to house prices. I estimate the current ratio at 4.5 per cent, which is probably the lowest it has ever been. Consider then the following scenario:
- House prices continue to rise in real terms but at a more moderate rate;
- Stocks also continue to provide real returns, though not quite as high as in recent decades, given the dampening effect of an aging population;
- Neither house prices or stocks suffer a crash;
- Mortgage rates gradually rise from their current lows but remain on the low side by historical standards;
- Inflation remains low;
- The ratio of rents to prices remains in the range between 4.5 per cent and 5 per cent.
Based on this scenario, renting is likely to remain the better option. Whether the reward is great enough to warrant the risk of a disastrous outcome is a different matter. In particular, consider the negative net worth of renters who had retired in the 1990s. Unlike the homeowners, their retirement security would have been in serious jeopardy. If the ratio of rents to house prices rises above 5 per cent, buying wins hands down.
Frederick Vettese is an actuary and the author of Retirement Income for Life. The second edition will be released on Oct. 20.