The future economic implications of the corona crisis are uncertain. If the current economic forecasts turn out to be correct it is expected that the COVID-19 pandemic will have a negative impact on the housing market due to lower household income and more uncertainty regarding future earnings. While the COVID-19 crisis is unique, there have been pandemics and epidemics in the past. In this article, we put the current pandemic and its potential effects on the housing market into a historic perspective.
This article previously appeared in our quarterly Dutch Housing Market Update. Want to receive these comprehensive, qualitative reports in your mailbox every quarter? Then sign up here.
In a recent study, Francke and Korevaar (2020)1 provide insights into the effects of historical epidemics in the housing market: they study multiple cases including the Paris cholera outbreaks f the 19th century and the Amsterdam plague outbreaks of the 16th and 17th centuries. Some of theplague epidemics killed 10% of the population of Amsterdam while the cholera epidemics killed about 2% of the population of Paris. The authors use historical housing transaction and rent data to construct price indices. They combine these indices with mortality data and approximations of outbreak starting dates, such that they can study the developments of the housing market around the outbreaks.
Similar studies have been done on the effect of outbreaks on the housing market. Ambrus et al. (2020) studies the 1854 London broad street cholera outbreak and its effects on rental prices and urban poverty and finds persistently lowered rental prices. Wong (2008) studies the impact of the 2003 SARS outbreak in Hong Kong on local house prices and finds that the outbreak caused a small decline in property prices of about 1.5%. Both studies, however, look at single cases of an outbreak, while Francke and Korevaar (2020) look at effects in the cross-section of epidemics as well as over time. They find that the outbreaks studied had a sizable downwards facing effect on aggregate house prices, but a less sizable effect on rents. The house price declines studied, were typically largest in magnitude during the first six months of the epidemic and in the most affected areas. Interestingly, these price declines were temporary as a swift recovery ensued after each crisis had ended.
Studies find that house price declines were largest in the first six month of an epidemic and that declines were temporary followed by a swift recovery
Impact of outbreaks on property prices
More specifically, they estimate the effect of each outbreak on property prices before and after the outbreak, controlling for variables such as consumer prices, wages, interest rates, and price growth three years around the outbreak. They document an average decline in house prices of about 5.5% per year during an epidemic. After the epidemic has subsided there is a decline of about 4.1% the following year. For rents, the effect sizes are 2.9% and 2.4% respectively. These effects remain robust, even when controlling for the variables mentioned. The researchers suspect that the smaller effect on rent prices compared to housing prices is due to the fact that rent prices are determined by supply and demand for local housing, whereas house prices are also affected by demand for capital and risk on the part of investors. They find this explanation appropriate as buy-to-let investors owned most properties in both cities.
When looking more specifically at transaction-level data for Amsterdam, they estimate the immediate response of house prices to an outbreak, controlling for annual price trends. They find that properties transacted within six months after the outbreak realize about 13% lower prices. In later months the effect is negligible or even positive.
They take a specific look at the Paris cholera outbreaks of 1832 and 1849 to investigate whether heavily infected neighbourhoods experience larger declines. When controlling for annual price trends, they find that a doubling of cholera mortality reduces neighbourhood-level house price growth following the epidemic by about 10%, but that this decline reverses quickly. The result for rents follows their general findings in the sense that the impact is smaller.
The impact on rents was smaller than on house prices
One of the limitations of their analysis it that, outbreaks are infrequent and therefore data availability is limited. It is therefore difficult to separate causal effects from underlying time trends, as the limited amount of observations used in the analysis could lead to measurement errors. Furthermore, the cases that they study are likely different from the situation we find ourselves in today. The cases studied could be characterized as epidemics while this time we are experiencing a global pandemic. The fact that we live in a globalized world facilitates the spreading of a virus such that a larger part of the world will be affected by the infection. The global interconnectedness of supply chains means that there are widespread negative effects when parts of these chains are disrupted. In later stages, a resurgence of the virus could lead to economic lockdowns once more, leading to wide spread negative implications for economic activity.
The authors state that the decline in housing and rent prices in the cases that they studied were due to the excess mortality itself as well as migration out of the infected areas. We do not expect these factors to play a significant role in the current pandemic in the Netherlands. In today’s environment, the declines in house and rent prices would more likely be due to declining economic activity as a consequence of the lockdown measures imposed upon by the government as well as a steep decline in overall consumer and investor confidence. The magnitude and duration of these effects mostly decided upon by how well businesses and consumers can deal with them as well as the effectiveness of government policy.
(1) Francke, Marc and Korevaar, Matthijs, Housing Markets in a Pandemic: Evidence from Historical Outbreaks (April 16, 2020). Available at SSRN: https://ssrn.com/abstract=3566909 or http://dx.doi.org/10.2139/ssrn.3566909
Dynamic Credit Partners Europe B.V. (‘Dynamic Credit’) is a registered investment company (beleggingsondernemingsvergunning) and a registered financial service provider (financieel dienstverlener) with the Dutch Financial Markets Authority (Stichting Autoriteit Financiële Markten). This presentation is intended for informational purposes only and is subject to change without any notice.The information provided is purely of an indicative nature and is not intended as an offer, investment advice, solicitation or recommendation for the purchase or sale of any security or financial instrument. Dynamic Credit may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented herein. Dynamic Credit cannot be held liable for the content of this presentation or any decision made by a third party on the basis of this presentation. Potential investors are advised to consult their independent investment and tax adviser before making an investment decision. An investment involves risks. The value of securities may fluctuate. Past returns are no guarantee for future returns.