Just released is a Canadian Economic Analysis Department working paper “On the Welfare Cost of Rare Housing Disasters“, which shows that in a significant housing downturn the welfare costs are large and this risk varies considerably across age groups, with a welfare cost as high as 10 percent of annual nonhousing consumption for the old, but near zero for the young. Given the risks to the housing sector in Australia at the moment, this is potentially important and applicable research.
Since the early 2000s, house prices have increased significantly in Canada.
This ongoing housing boom has become an important consideration for the conduct of monetary policy and financial regulation, since currently high levels of house prices are potentially increasing the risk of a large housing market correction, which could have an adverse effect on the economy. This paper investigates the likelihood and magnitude of housing market disasters, and the value of limiting this disaster risk for the Canadian economy.
The study will be useful for both economic researchers and policy-makers to better understand the macroeconomic implications of this important market risk. This paper estimates the unconditional probability and the size of housing market disasters using the cross-country housing market experiences of twenty OECD countries.
I find that in a given OECD country, housing market disasters – defined as cumulative peak-to-trough declines in real house prices of 20 percent or more – occur with a probability of 3 percent every year, corresponding to about one disaster occurring every 34 years. A housing disaster on average lasts about 6.4 years, and house price declines range from 25 to 68 percent, with an average decline of 34 percent.
This paper quantifies the welfare impact of the housing disaster risk in an overlapping generations general equilibrium housing model. The analysis yields the following two main findings. First, despite their statistical rarity, the aggregate welfare cost of housing disasters is large, since Canadian households would willingly give up around 5 percent of their non-housing consumption each year to eliminate the housing disaster risk. The sizable welfare cost is due to the large wealth loss during the long-lasting recessions triggered by housing disasters. Second, the welfare evaluation of this risk varies considerably across age groups, with a welfare cost as high as 10 percent of annual nonhousing consumption for the old, but near zero for the young. This asymmetry stems from the fact that, compared to the old, younger households suffer less from house price declines in disaster periods, due to smaller holdings of housing assets, and benefit from being able to buy homes at the resulting lower house prices in normal periods.
The main findings from the model are twofold. First, despite their rarity, the aggregate welfare cost of housing disasters is large, since Canadian households would be willing to give up around 5 percent of their non-housing consumption each year to eliminate the housing disaster risk. Compared to the no-disaster economy, the presence of this disaster risk has two opposite welfare effects on households. On the one hand, due to a wealth effect, a realized housing disaster leads to a long-lasting economic recession. The loss in housing wealth once a disaster occurs reduces the aggregate household savings and thus the capital supply. As a consequence, the interest rate goes up and the fi rm cuts back its investment, leading to declines in wages, output and consumption. On the other hand, due to a substitution effect, a non-trivial disaster probability results in risk-averse householdsresource reallocation from the housing sector to the production sector in normal periods. This lowers both house prices and the interest rate, with declining borrowing costs leading to higher investment, output and consumption. However, due to diminishing marginal utility of consumption, the welfare gain from this resource reallocation in normal periods is dominated by the welfare loss from large recessions triggered by housing disasters, explaining why the society is willing to devote a sizable amount of resources to eliminating this disaster risk.
The second major finding is that the welfare evaluation of the housing disaster risk differs considerably in magnitude across age groups, with a welfare cost as high as 10 percent of annual non-housing consumption for the old, but near zero for the young. This asymmetry is mainly due to the hump-shaped pro le of life-cycle housing consumption, with older households typically holding more housing assets than the young. In disaster periods, declines in house prices favor the young, who purchase houses at depressed prices, but hurt the old, who rely on house sales to finance non-housing consumption. In normal periods, younger households also benefit more than the old from purchasing houses at lower cost, thanks to the resource reallocation from the housing sector to the production sector as discussed above. Therefore, younger households are less averse to the presence of the housing disaster risk than the old.
Note: Bank of Canada working papers are theoretical or empirical works-in-progress on subjects in economics and finance. The views expressed in this paper are those of the author. No responsibility for them should be attributed to the Bank of Canada.
