MLS® Home Price Index explained
The MLS® Housing Price Index (HPI) is a measure of real estate prices that provides a clearer picture of market trends over traditional tools such as raw average or median prices.
The raw average price is obtained by dividing the total dollar volume of sales by the number of sales.
To obtain the median price, all of the sales prices are arrayed in ascending order, and the middle price is taken. In the case of an even number of sales, the median is the highest price in the lower half of the group. If there is an odd number of sales, the midpoint sale is taken as the median.
The MLS® HPI is modelled after the Consumer Price Index, which measures the rate of price change for a basket of goods and services. A basket is a combination of goods and services that Canadians buy most such as food, clothing, transportation, etc.
Instead of measuring goods and services, the MLS® HPI measures the rate at which housing prices change over time taking into account numerous features about the types of homes sold.
The problem with raw averages
Before the original HPI was introduced in 1996, REALTORS® and the public relied on monthly raw average price statistics to understand trends in housing prices.
Raw averages, however, can be very misleading since the quantity and quality of the properties sold in any given area change over time for any number of reasons. As a result, average prices can fluctuate dramatically, making the housing market appear unstable.
To demonstrate this point, let’s look a couple of examples of how average prices are affected by various changes in the composition of sales.
Example 1: How raw averages are affected by the composition of sales
| Year 1 ($) | Year 2 ($) |
|---|---|
| 1. 139,000 | 1. 139,000 |
| 2. 145,000 | 2. 145,000 |
| 3. 230,000 | 3. 230,000 |
| 4. 265,000 | 4. 265,000 |
| 5. 290,000 average | 5. 290,000 median |
| 6. 320,000 | 6. 320,000 |
| 7. 365,000 | 7. 365,000 |
| 8. 425,000 | 8. *545,000 |
| 9. 480,000 | 9. *580,000 |
| Total $2,659,000 ÷ 9 sales = $295,444, which is the raw average | Total $2,879,000 ÷ 9 sales = $319,888, which is the raw average *price change from Year 1 |
In this example, the raw average price increased by 7.7 percent while the median price stayed the same. This shows that the composition of sales at either end of the price spectrum affect the raw average but can leave the median price unchanged.
Example 2: How median prices are affected by price changes
| Year 1 ($) | Year 2 ($) |
|---|---|
| 1. 139,000 | 1. 139,000 |
| 2. 145,000 | 2. 145,000 |
| 3. 230,000 | 3. 230,000 |
| 4. 265,000 | 4. *290,000 |
| 5. 290,000 median | 5. *320,000 median |
| 6. 320,000 | 6. *335,000 |
| 7. 365,000 | 7. *395,000 |
| 8. 425,000 | 8. *400,000 |
| 9. 480,000 | 9. *405,000 |
| Total $2,659,000 ÷ 9 sales = $295,444, which is the raw average | Total $2,659,000 ÷ 9 sales = $295,444, which is the raw average *price change from Year 1 |
In this example, the raw average stayed the same while the median average increased by 9.4 percent. This shows that the composition of sales in the mid-range of the price spectrum can affect the median price, but can leave the raw average unchanged.
Neither of these price measurements take into account the changes in buying patterns, nor important distinguishing characteristics of the homes sold. In year one, luxury homes in the region may be more popular and sell more frequently; the following year, more modestly priced homes may be popular.
Both methods of price tracking can have the effect of over- or under-estimating the market price.
Defining the typical home
The MLS® HPI is a more stable price indicator than raw average or median prices, because it tracks changes of "typical" homes, and excludes the extreme high-end and low-end properties.
Typical homes are defined by various quantitative property attributes (e.g., above ground living area in square feet) and qualitative features (e.g., proximity to shopping, schools, transportation, hospitals etc.). The MLS® HPI uses these features to estimate a more accurate price of the “typical” home in Greater Vancouver, based on how these features impacted the value of similar homes which sold recently.
These features together become the "benchmark" house, townhouse or apartment in a given area. A benchmark property is designed to represent a typical residential property in a specific MLS® HPI housing market, such as Richmond or North Vancouver.
For example, perhaps the basket of features for a typical home in a given community are defined as being a 10-year-old, 3-bedroom house without a panoramic or ocean view on a 7,200 sq. ft. lot, with 8 rooms, 2 bathrooms, a fireplace, a 1-car garage and is close to schools. A benchmark price for this home can be created from the individual dollar values ascribed to each of these features, based on recent data of similar properties that have sold.
More information about the MLS® HPI can be found at the following link to the Canadian Real Estate Association (CREA):
CREA | MLS® Home Price Index (HPI)
Note: The MLS® HPI offers only a benchmark in which to track price trends and consumers should be careful not to misinterpret index figures as actual prices. Benchmark properties are considered average properties in a given community and do not reflect any one particular property.