An explanation of CHBA's methodology for the HMI is below. For a more comprehensive overview, click here.
Conducted on a quarterly basis, the CHBA survey asks an expert panel to rate market conditions for the sale of new homes at the present time and in the next six months as well as the traffic of prospective buyers of new homes. Each variable allows the respondent to rate conditions as either “Good”, “Fair” or “Poor” or in the case of the traffic of prospective homebuyers “High/Very High,” “Average,” “Low/Very Low.” Each of these three variables creates a component index and the HMI is an average of the three component indices:
Component Indexᵢ = (% of Good responsesᵢ - % of Poor responsesᵢ +100)/2
HMI = 0.6xPresent + 0.1xFuture + 0.3xTraffic
The final number is weighted to ensure it is representative of CHBA’s membership by geographical location. The provincial weights leveraged 2020 Statistics Canada housing starts data for the breakdowns of single and multi-family markets to create the proportions of the single-family and multi-family markets in each province. This allows CHBA to create an accurate provincial representation while also taking into consideration that each province has very different single-family and multi-family proportions. Each local Home Builders’ Association (HBA) is also weighted to ensure proper representation by builder size (I.e., small/medium and large). In some cases, where local breakdowns for HBAs is not feasible, those HBAs are not weighted. The provincial weights are developed in a way to ensure the sample size is representative of the overall population of CHBA's membership of 2,861 builders. The special questions presented above are only weighted at the provincial level.
For the Q1 results, an online survey of 248 builders was conducted by CHBA from February 24 to March 17, 2022, using CHBA's panel of builders. Some builders elected to speak to both the single and multi-family markets, resulting in 340 unique responses. 43 local HBAs participated in this survey. The full sample has a margin of error of 5% at a 95% confidence level.
CHBA’s HMI is modelled on that of the National Association of Home Builders (NAHB) in the US. The NAHB-Wells Fargo Housing Market Index is based on a survey of single-family builders that the NAHB has been running every month since 1985.
For this year, CHBA is leveraging the NAHB's weights as the basis for CHBA’s HMI weights (see formula above). This is to ensure that CHBA is placing more weight on the Present conditions question. As CHBA collects more data, it is planning on developing weights for each of the component indices discussed above that will allow us to maximize the correlation with Canadian housing starts six months into the future. CHBA did not create the weights this year due to the need for historical data (which will only be possible as the HMI continues moving forward) and the need to rely on housing starts forecasts, which may not be accurate. In order to maximize the HMI’s ability to predict single-family and multi-family starts six months in the future, CHBA be using a longer time horizon of its HMI data to identify and examine a substantial trend in the data.
The panel is made up of builders from constituent associations to ensure representativeness across Canada. Working together with the association’s local and provincial Executive Officers, CHBA was able to build this expert panel that is reflective of membership in each region. CHBA is planning on refreshing the panel once a year. The HMI is weighted to ensure the panel is representative by both geographical location (i.e., national, provincial and local) and size (i.e., small, medium and large). For some smaller home builder associations (HBAs), the sample is not weighted. In those cases, CHBA leverages the HBA’s expertise to ensure the sample is representative of the region. The panel also has a balance of single-family and multi-family builders that reflect the conditions of each local market in Canada. Single-family includes single-detached homes, semi-detached homes and row (townhouse) homes. Multi-family includes stacked townhouses, duplexes, triplexes, double duplexes and row duplexes, and low and high-rise apartment buildings.
Q1 Survey Demographics