Real estate markets are inherently local in nature. The value of a property is informed by the sale of similar nearby properties in the recent past. Thus, house price changes in one city may have no relevance to prices in another.
Recent research reinforces the notion that for housing markets, the extent of local may be restricted to the neighbourhood level. Housing values in a mid-income neighbourhood will not necessarily respond to fluctuations in prices in a ritzy part of the town.
Whereas homebuyers and sellers, and real estate agents already work with markets at the neighbourhood level, some segments of the industry see housing only at the national or, at best, regional level.
In a world awash with data, cheap data storage, ubiquitous computing and advanced analytics, providing housing insights at the neighbourhood level should no longer be an insurmountable challenge. In fact, researchers at the U.S. Federal Housing Finance Agency have demonstrated how house price indices can be generated at the postal code level.
Writing in the journal Real Estate Economics in 2019, Alexander Bogin and co-authors showcase a new series of house price indices for 914 cities, 2,716 counties, 18,053 five-digit postal codes and 54,901 census tracts. For each local housing market, they developed an annual time series of indices covering transactions over four decades.
The authors have gone beyond what most indices were able to do in the past. Consider Canada, where the oft-cited price indices include the Teranet-National Bank (TNB) House Price Index and the Canadian Real Estate Association’s (CREA) MLS Home Price Index. These indices come in two flavours. The national index is a composite of a subset of urban housing markets. The local indexes are based on the transactions recorded in large urban housing markets.
The city-wide index is a useful metric to know about the trends in the broader regional housing market. It will identify times where the housing market was up or down, thus helping the buyers and sellers in their decision-making.
However, as the cities have become larger over time and have amalgamated other towns to become city regions, a single index may not suffice. Both MLS and TNB indices report an index for Toronto, which has a regional population of more than six million. Housing and land prices vary considerably in Toronto. For instance, research by Murtaza Haider and Eric Miller in 1999 showed how housing prices for near-identical units were higher near downtown Toronto and declined with distance from downtown.
A neighbourhood level index will be of higher value to buyers and sellers because buyers search for a specific type of housing and their spatial search is usually confined to a select few neighbourhoods and not the entire city. Thus, house price fluctuations in neighbourhoods of interest matter more to buyers than what the price trends are in other parts of the city or in other cities.
The local price indices developed by Bogin and his co-authors generated insights that remained hidden from other aggregate indices. The authors found that house price appreciation in neighbourhoods near downtowns in large American cities was considerably higher than in places that were 10 miles away from the city centre.
Insights about the difference in house price appreciation are valuable to all, but more so for investors who would like to know what parts of the city offer higher rates of return on housing investments.
Local house price indices suggest that an increase in housing demand in areas with highly elastic housing supply, implying that new housing can be built readily in response to higher demand, will result initially in higher prices. However, over time and with new supply, prices return to their long-term trends.
The same is not valid for supply constrained places. Because of development restrictions, restrictive regulations, or expensive land prices, housing supply will not respond commensurately to shocks in housing demand. The authors point to places near downtowns, “where buildable sites are less available and regulation is presumably more onerous, a permanent demand shock can outpace supply responses, leading to price increases.”
Housing transaction data is available with municipal governments and local real estate boards. Statistics Canada has also been working on developing a housing transaction database. The markets would be served better if the existing data were put to good use leading to insights for informed decision-making. Local house price indices will be one such intelligent use of data.
Murtaza Haider is a professor of Real Estate Management at Ryerson University. Stephen Moranis is a real estate industry veteran. They can be reached at www.hmbulletin.com .
Copyright Postmedia Network Inc., 2019