To: The Hon. Sean Fraser, Minister of Housing Infrastructure and Communities of Canada
cc: The National Housing Council
From: Marc Francke, Lianne Hans, Matthijs Korevaar, and Sjoerd van Bekkum
Date: September 7, 2023
Re: Buy-to-Live vs. Buy-to-Let: The Impact of Real Estate Investors on Housing Costs and Neighborhoods
In many places across the world, housing has become increasingly unaffordable. Policymakers are trying to understand the reasons behind the surge and, more importantly, what can be done about it.
In Canada and elsewhere, attention has turned to rental investors. Could rising investor activity be driving higher housing costs, displacing first-time buyers, and changing neighbourhoods? And if so, what would be the consequences of regulating such activity?
To help answer this question, we studied the effects of a January 2022 Dutch law aimed at constraining investors from crowding out regular homebuyers. It allowed municipalities to effectively ban investors from buying affordable owner-occupied property with the intention of renting it out. Municipalities could decide for themselves whether and when to implement the buy-to-rent ban. They could also decide where to apply it, and below which predetermined tax value such conversions are banned. In most cities, these tax value limits were sufficiently high to cover the large majority of such transactions before the policy was introduced. The variation in policy coverage provides a unique field experiment to study the impact of investors.
In our paper, we present the first results of our evaluation of the policy. Our study covers the entire country, but we focus specifically on the city of Rotterdam, which was the first city to introduce the policy and the only major city to apply it only in specific neighbourhoods. By comparing subject neighbourhoods with adjacent neighbourhoods both before and after the policy, we can precisely pinpoint the policy effects.
We employed near-real-time administrative data on the residences of the entire Dutch population, their personal characteristics, incomes, and the owners of all these properties. This allowed us to track for every sale of owner-occupied property whether purchased by an investor, and whether it was subject to the ban. As in the US and Canada, most buy-to-rent investors are private individuals or smaller commercial investors.
In Rotterdam, we found the ban decreased the likelihood that a property would be bought by an investor by 23 percentage points: a reduction of more than 75 percent from the pre-policy level. First-time buyers filled most of the void. Results are similar at the national level: The policy was effective at replacing investors with owner-occupiers.
We then studied the effect on sale prices by comparing properties subject to the ban to similar properties not subject to it. Both within Rotterdam and across the country as a whole, there was no significant effect on house prices in the first year of implementation. We found some evidence of increased rental rates, in line with reduced listings, but this effect should be treated as suggestive given the limited nature of rent data.
The stability of house prices could be related to the fact that investors were able to buy properties without bidding higher prices compared to owner-occupiers. For example, most people buying houses need to make their bids conditional on receiving mortgage financing and frequently have restrictions on closing dates, while investor bids usually do not include such clauses. Another potential reason is that any reduction in investor demand was offset by higher demand from owner-occupiers. One motivation for the policy was that own-to-rent conversions were perceived to negatively affect neighbourhoods. In line with this, several resident groups in Rotterdam have lobbied to expand the policy to other areas. If a ban is perceived as an amenity, it might also increase demand from other owner-occupiers.
To study this sentiment in more detail, we exploited our detailed administrative data to see how investor ownership vis-à-vis owner-occupancy affected neighbourhood composition. The investor ban and the corresponding uptick in purchases by owner-occupiers resulted in the entry of residents with higher incomes.
The reason is that renters in investor-owned properties have substantially lower incomes than residents of equivalent owner-occupied property. Tenants are also younger, more likely to be immigrants, and much more likely to move within two years compared to owner-occupiers. Investor-owned properties also house more residents. These effects are found consistently across the Netherlands, although their magnitude differs across areas. (We should note that the Netherlands has a large income-tested social housing sector, which means low-income residents are likely still underrepresented in the Dutch private rental sector compared to other countries. As in North America, however, long waiting lists still force many lower income people into the private rental market.)
In short, the Dutch evidence suggests that banning investors has these effects because tenants and owner-occupiers have different characteristics even if they live in very similar properties. Restricting buy-to-let activity might help middle-income residents gain entry into homeownership, but this comes at the expense of lower-income residents who rely on private rental housing and generally have insufficient income to buy a house.
For Canada, it seems important to keep such consequences in mind when considering regulating rental investors. So far, our results also indicate that restricting investors does not improve affordability nor combat housing scarcity.
We want to end with two remarks. First, as the policy is recent, we continue to monitor it and will update our paper and findings as more data becomes available. Second, our study focused on the housing market effects of buy-to-let investment by for-profit investors. Investors could also affect local housing conditions by constructing and redeveloping areas – build-to-rent – and governments could help the affordability crisis making sizable investments in public housing. Our results do not speak to the effects of such investments. However, one crucial lesson can be drawn: Housing investments seem to have a direct effect on who lives where.
Marc Francke is Professor of Real Estate Analytics at the University of Amsterdam; Lianne Hans is a Researcher at Kadaster; Matthijs Korevaar is Assistant Professor of Finance at the Erasmus School of Economics where Sjoerd van Bekkum is Professor of Finance and Economics.
To send a comment or leave feedback, email us at blog@cdhowe.org.
The views expressed here are those of the authors. The C.D. Howe Institute does not take corporate positions on policy matters.
