Appraisal Institute of Canada
2009 – Volume 53 – Book 4
Do Restrictions on Renting Increase Condominium Values?Tsur Somerville
, Chris Kay
, Seung Dong You
Evidence from Vancouver*
By Tsur Somerville‡, Chris Kay, and Seung Dong You
Property owners, in association with one another, can choose to place restrictions on their individual property rights. Commercial property owners may choose to do this as part of a business improvement district; residential owners may act through their homeowner’s association or by vote of a strata council in a strata title community1. An important question for valuation is whether the imposition of restrictions affects the market value of the units. The research presented here studies the effect of one particularly strong restriction: strata by-laws that forbid owners of condominium units in the strata from renting their units.
What effect rental restrictions should have on prices and values is not clear. Restrictions reduce a potential investor’s return, making a unit in a building that forbids rental of lower value to investors2. This will push down market values by reducing demand. On the other hand, owner-occupiers might prefer a building with rental restrictions, such that they would pay a premium. First, if renters are perceived to be risky or undesirable, then restricting them makes the building more attractive to owner-occupiers. Second, if there is no benefit, why would owners vote for these restrictions, which require a supra-majority of 75% to pass?
Transaction data on condominiums in Vancouver, BC and the adjacent suburbs of Richmond and Burnaby is used to measure the effect of these restrictions on unit values. This issue is of particular importance in Vancouver. Apartment-type buildings made up 65% of new construction in Vancouver in 2007-08 and 50% of starts between 1996 and 2008, compared with 45% and 28% respectively in Toronto. Unlike other cities, in Vancouver, these starts are almost entirely strata titled buildings sold to individual buyers rather than rental buildings owned by a single individual or entity3. Investors make up a substantial share of the owners of condominium units in the Vancouver area. In the city, 35% of condominium units appear to be investor owned4. The figure is lower in Vancouver’s suburbs, but still significant at 22%. The absence of new purpose-built rental construction and the decline of the existing stock of purpose-built rental units means that investor-owned condominium units, along with basement suites, are and can be expected to be an important part of the rental housing stock in the Vancouver area. Restrictions on the ability of investors to rent their units can be expected to not only affect values, but the ability of the Vancouver area to supply rental housing options.
While this is the first study we know of to examine the effect of restrictions on leasing on condominium values, there is a body of research that has attempted to identify the effects of collective property rights restrictions through private covenants on property values. The main body of this work has looked at gated communities and ‘homeowner associations’ (HOAs), in part, because of the criticism the gated community has received from planners and sociologists6. Among economists, the objective of research has been to identify the magnitude of any effect on value for a home being in such a community and the source of these effects, either from enhanced security, reduced traffic flow, or restrictive covenants that limit the risk of actions by one household having a negative effect on the value of neighbouring houses. General effects for the presence of an HOA range from Agan and Tabarrok’s (2005) finding of a 5.4% premium for houses in a community with an HOA to Hughes and Turnbull (1996), who find the price premium for being in a community stricter covenants is between 2% and 6%. The gated aspect itself adds 6% to value in Bible and Hsieh’s (2001) study. The findings in this research appear to be quite sensitive to the location of the analysis. Two other studies have found much larger effects: Pompe (2008) finds an 18.6% premium for gated communities in beach resorts, while LaCour-Little and Malpezzi (2001) find that in a St. Louis neighbourhood , house prices are 26% higher in a gated community – 17 percentage points from the presence of an HOA and nine percentage points from the gate. The precise relevance of these studies of single-family housing, typically in more suburban locations for the valuation of urban condominiums is not clear.
Two studies have looked explicitly at restrictions in condominium units, though not at the restrictions on leasing that is the focus of our work. Allen (1997) compares condominiums with and without age restrictions in Florida and found that buildings that limit the occupancy to older residents received a 20% premium to similar un-restricted units. Cannaday (1994) presents a simple model that suggests limited constraints are more valuable to owners in a building than stricter constraints. Examining a set of condominiums in Chicago, he finds that allowing cats only raises property values by 5.5% relative to forbidding pets entirely, while allowing dogs, either small or large, lowers values. Both of these effects are surprisingly large, given that the formal ‘cost’ of imposing restrictions is quite low, and that, if the premiums are so high, we would expect more units to be ‘supplied’ via strata corporation by-laws.
Our data analysis starts with the universe of strata-titled condominium units in the cities of Burnaby, Richmond and Vancouver, BC. Limiting our analysis to buildings with at least five units gives us a sample of 1,781 buildings with 107,611 individual units, which is approximately 1/4 of the housing stock in these three cities. We received transaction price and unit characteristic details for these units from the British Columbia Assessment Authority (BCAA). When a strata-title property is listed on the Real Estate Board of Greater Vancouver’s (REBGV) multiple listing service (MLS), the listing agent has the option of indicating whether leasing of the unit is allowed by the strata corporation or not. We use these records for property listings in 2007-2009 to determine whether or not unit owners are allowed to rent their units. This limits our data to approximately half of the universe of buildings (897 strata plans), but this includes about 70% of the units (72,071 individual units).
Rental restrictions are not trivial. Approximately 13% of the units are in buildings that forbid the leasing of units in the building. There is a clear geographic pattern: 6.6% of units are restricted in Vancouver, compared with 14% in Richmond and 26% in Burnaby. Strata corporations with restrictions are smaller (a mean of 62 units) compared with 85 units in those that do not forbid leasing. Not surprisingly, the share of units owned by investors in buildings with restrictions is lower than those without: 15% compared to 26%. The two are related, as the largest 25% of buildings have a 31% investor share and only 11% of these have rental restrictions, as compared with a 26% investor share and 20% with restrictions for the smallest quartile of buildings.
To estimate the effects of these restrictions on values, a hedonic analysis is performed on transaction prices. In a standard statistical appraisal approach, a unit’s transaction price is estimated as a function of unit characteristics, the year and quarter of the transaction, whether leasing is allowed, and controls for each of the BCAA-designated assessment neighbourhoods7. Transactions from between January 2007 and April 2009 are used and the analysis is conducted combining all the transactions, by year, and by city.
We find that forbidding rentals in a building lowers the price owners receive when they sell by between 2.7% and 3.3%. This result is robust and consistent across the different types of statistical analysis conducted. The one exception is that units in strata buildings in the West End of Vancouver, an area with very high numbers of renters and rental buildings, actually receive a premium of 10%. It would not be surprising that, in an area where renters are the dominant tenure, being able to differentiate your product can yield a positive effect on unit values. Excluding the West End from the more general analysis strengthens the negative effect of restrictions on price: to 3.3% to 5.4%.
Along with price, unit owners can be expected to care about liquidity: how easy it is to sell their units. We are unable to formally test this effect with the data we have. We know if a unit is sold, but not how long it took to sell, or whether a seller withdrew the unit from the market. We do analyze the effect of restrictions on whether a unit sells or not. We find that rental restrictions will raise the probability that a given unit is sold by 1.2 to 2.8 percentage points. Given, that restrictions are associated with a decline in prices, this results is more consistent with owners of units in buildings with restrictions being more likely to sell, than an actual increase in liquidity.
This study attempts to achieve two goals: a very narrow objective of giving valuators and property owners a yardstick for assessing a certain type of owner-imposed restrictions on the ability of individual property owners to exercise their rights: the restriction preventing the leasing of condominium units; and a second more general objective to increase our understanding of the patterns of these restrictions and identify aspects of their potential impact on housing markets. The results of the former are quite clear: a general 3% decrease in property value. This effect is higher, closer to 4.5% in areas where rental is not the dominant form of tenure. Interestingly, the restrictions can have a positive effect in situations where the restrictions limit a clearly perceived and likely use. As to the second objective, the results are more general, but they do highlight the clear geographic variation in the application of restrictions; restrictions are more likely to occur in older buildings with more owner-occupiers and fewer investors.
Strata title developments are typically condominiums, but can be townhouse or single-family detached and non-residential developments.
2 The Strata Act in BC allows existing renters in a building that passes a strata by-law to forbid rentals to be grandfathered in perpetuity, and owners have a one year window after a grandfathered tenant has left, during which the unit may still be rented.
3 CMHC Marketwire 12/13/07 http://www.marketwire.com/press-release/Canada-Mortgage-And-Housing-Corporation-802631.html.
4 We define a unit as investor-owned if the property tax assessment is sent to an address other than the unit itself. If an investor has the assessment sent to the unit, we are likely to underestimate the number of investors.
5 CMHC estimates this to be 20% of the rental stock in the Vancouver metro area (see CMHC, Rental Market Report: Vancouver and Abbotsford CMAs, Fall 2008). For a discussion of the decline of the purpose-built rental stock in Vancouver see http://thetyee.ca/Views/2009/03/04/RentalHomes/.
6 See Blakely and Snyder (1997) as an example.
7 For residential properties, there are 40 in Burnaby, 37 for Richmond and 32 in Vancouver (including areas near the University of British Columbia). For Burnaby and Richmond, a ‘neighbourhood’ may be specific to a type of property.
Agan, Amanda, and Alexander Tabarrok. ‘What Are Private Governments Worth?’ Regulation. 2005, (Fall): 14-17.
Allen, Marcus T. ‘Measuring the Effects of ‘Adults Only’ Age Restrictions on Condominium Prices.’ Journal of Real Estate Research. 1997, 14 (3): 339-46).
Bible, Douglas S., and Chengho Hsieh, ‘Gated Communities and Residential Property
Values.’ Appraisal Journal. 2001, April: 140-46.
Blakely, Edward, and Mary Gail Snyder. Fortress America. 1997. Washington, D.C.: Brookings Institution.
Cannaday, Roger E. ‘Condominium Covenants: cats, Yes; Dogs, No.’ Journal of Urban Economics. 1994, 35: 71-82.
Hughes, William, and Geoffrey Turnbull. ‘Uncertain Neighbourhood Effects and Restrictive Covenants.’ Journal of Urban Economics. 1996, 39: 160-72.
LaCour-Little, Michael, and Stephen Malpezzi. ‘Gated Communities and Property Values.’ 2001. CULER working paper. University of Wisconsin-Madison.
Pompe, Jeffrey. ‘The Effect of a Gated Community on Property and Beach Amenity Valuation.’ Land Economics. 2008, 84 (3): 423-33.
* This work was made possible with the generous support of AIC, AIC British Columbia Chapter (AICBC), The Real Estate Foundation of BC, and the Real Estate Institute of BC (REIBC). We thank the BC Assessment Authority and the City of Vancouver for providing data. This article is part of a larger effort on the effects of rental restrictions in partnership with Michael Alexander of Smart Growth BC. The research was conducted under the auspices of the UBC Centre for Urban Economics and Real Estate at the Sauder School of Business.
‡ Real Estate Foundation Professorship of Real Estate Finance, Sauder School of Business, University of British Columbia, [email protected]