I should warn you that this is a little bit of statistical geekdom coming your way.
This comes from the REBGV to explain how the Board’s Vancouver real estate statistics are measured and derived. They do not use averages as I do on the weekly stats for the very reason that I have discussed before (wild swings and skews). Instead, they have derived a benchmark measurement.
I still think there is value in averages as it does paint a picture in the overall trend. A change was recently made with the addition of the “4 week average” so that you can see if the past week was greater or worse than several weeks combined. Fairly consistent dropping of the average price does not determine your specific property’s value but it does point to the trend of the general market.
From the REBGV:
Are housing prices trending up or down? The state of home prices in Greater Vancouver is a much discussed topic at the moment. Last month several organizations issued reports examining the current and future direction of home prices in our market. The BCREA Fall Housing Forecast recently reported that average home prices for residential properties, meaning detached, townhome and condo units, in Greater Vancouver are expected to
increase three per cent this year compared to 2007. On the heels of this release, the Canada Mortgage and Housing Corporation (CMHC) similarly projected that residential housing prices across Greater Vancouver will increase three per cent by the end of 2008 compared to the previous year.
These numbers seemingly conflict with the figures the Board has released in recent months. According to the REBGV statistics release for October, residential home prices, as calculated by the Housing Price Index (HPI), have declined 8.8 per cent between May and October 2008, eliminating price
gains witnessed in the first quarter of 2008.So what accounts for the discrepancy? The prices in the aforementioned BCREA and CMHC forecasts cite averages, perhaps the most commonly
referenced housing price measurement. In contrast, the Board uses
benchmarks prices (see definitions below) as its primary indicator of home values.
As of Oct. 31, average residential home prices had increased five per cent in Greater Vancouver over the last 12 months to $599,068 from $569,903. Benchmark prices, however, indicate a 3.9 per cent decline to $518,668 from $539,703 over the same period.Greater Vancouver has experienced unprecedented growth in home values the last five years. Both benchmark and average price data tells this story, though the numbers differ.
Between December 2003 and February 2008, the benchmark price of a detached home in Greater Vancouver increased 69 per cent from to $761,342 from $449,190. The average price of detached homes over the same period registered a 94 per cent increase to $920,644 from $475,087. For condominiums, benchmark prices increased 82 per cent to $387,032 from $213,140, while the average price increased 81 per cent throughout this five-year span to $424,839 from $233,733.
Whether it’s a seller’s or buyer’s market, the Board aims to convey the most accurate and current housing market information to members and the media. Following extensive consultations with economists more than a decade ago, the Board identified benchmark prices, produced by the HPI, as the most accurate measurement of home property values over time.
Similar to the Consumer Price Index that tracks inflation and measures the rate of price change for a basket of goods and services including food, clothing, shelter, and transportation, the HPI measures the change in the price of housing features such as lot size, age of the home and neighbourhood etc. These features become the composite of the ‘typical house’ in a given area.
Average prices are calculated by dividing the total dollar volume generated from home sales by the number of sales in a given period. Averages, however, can be misleading since the quantity and quality of properties sold in a given area change over time. As a result, average prices can fluctuate considerably, making the housing market appear unstable. The HPI solves this problem by
pricing a constant quality and typical property over time.Whichever measurement is used, regional housing reports only provide a broad outlook. Viewing the “housing market” at large offers only a snapshot of the collective activity within a region. Prices today are changing by area. Establishing a price depends on the property type and neighbourhood. Ultimately real estate is local and REALTORS® offer the expertise that allows clients to identify opportunities that match their local needs.
Vancouver, BC 
Getting even geekier, the “benchmark” price uses, I think, hedonics to try to estimate price by giving relative value to rooms, lot size, number of bathrooms, etc. I believe it does not “quality” adjust — whether there is cherrywood or laminate flooring or high end Delta faucets for example. The US measures such as the Case-Shiller index do not account for quality either.
For example say you buy a standard 4 bedroom 2 bath bungalow, do no upgrades and sell it 4 years later (in early 2008 LOL) for a 50% profit. Both Case Shiller and the benchmark will track this property roughly the same. However let’s say you completely renovate the property with state-of-the-art interior and you sell for a 60% profit. Again, both measures will track the property the same — 60% — however this second case is overstating gains since the neighbour that did no renovations will look at the benchmark and assume his property is worth 10% or so more than it really is. Also your true gains are net of renovation expenses which are roughly, say, 10%.
In large markets over a long period the luxury upgrade offset should be cancelled since there is a large distribution of properties at various vintages. But it means that if you choose to lag the average house in terms of maintenance and upgrades your price gains will be lower than the benchmark. It also means you need to spend on maintenance to have your property keep up with the benchmark. In a boom cycle it could have been that renovations were more common and the benchmark would have increased more than true “same” properties. This is, in fact, part of the reason why prices are “sticky” on the way down: headline prices do not account for boom-time renovations.
I would love for REBGV to release its HPI formulas but I’ve inferred enough from their high level explanations to have confidence in what they are trying to do. Certainly better than the average but some just can’t take off the tinfoil hat. Sigh.
No, there’s simply no way to account for upgrades. How could they? More importantly what be would the value of doing so? I’d say that the point of measuring the HPI Benchmark is simply to overcome the wild swings that are caused significantly different properties on the average.
No, the value in this, just as the value in the various stock market indexes, be it Dow, S&P, or TSX 100, is simply to show an overall trend. Averages can show up or down X%. HPI can do the same with different numbers. Only an appraisal can point to what your specific property will sell for up or down from before and then only the market and the Seller can determine what its value actually is at that point in time, upgrades or not.
As an aside, I’d say that upgrades right now will not add any value to your home… they will merely improve its salability (and that is a very good thing). In a stable or rising market, they do increase the value. Not in a falling market.