I just finished appraising three typical, but different, properties in downtown Gladstone, Oregon. The three Gladstone properties are a 1940’s 1,200 square foot ranch, a 1940’s duplex, and a 1960’s 2,900 square foot house. Although all properties have a similar location and value date, all have different market analyses, and different market conclusions.
The Gladstone, Oregon properties highlight market conditions and how they relate to individual property types. Many Oregon professionals including the media, real estate agents, and appraisers looking for real estate market statistics will go straight to the RMLS (Realtor Multiple Listing Service) Market Action. While this report is a good source of information, I will show you why the numbers are misleading.
The RMLS Area Report has a “Percent Change” statistic listed for each market area and region. This number is used by many appraisers for time adjustments and the number is a favorite quote for real estate agents and the media. The Percent Change is calculated “based on a comparison of the rolling average sale price for the last 12 months (5/1/09-4/30/10) with 12 months before (5/1/08-4/30/09).” The following is a table of data showing how the RMLS produced its Percent Change statistic for Area 145 (includes Gladstone, Oregon) in the April 2010 report:
Percent Change = -0.0948
Percent Change = -9.5%
Real estate professionals and the media often present this statistic as though prices dropped 9.5% in Area 145 over the past year. The problem is that this might not be true for most properties. The following is a graph showing the same average sales price data, used to develop the Percent Change statistic, charted over the total two year time period:
The graph shows that prices didn’t just drop 9.5% as they suggest over the last year. My interpretation of this chart alone is that prices declined sharply thru 2008, were stable with bumps thru the end of 2009, and may be beginning to decline again as shown by the black trend line. I think that the drop in 2010 is skewing, as I will discuss.
The RMLS Area 145 has a mix of all types of residential properties and price ranges. Area 145 includes the Oregon cities of Milwaukie, Gladstone, Happy Valley, Clackamas, and Damascus. Even within the City of Gladstone, Oregon, there is a diverse mix of properties with different sets of buyers and sellers who are driven by different economic and social circumstances. It has been widely reported that larger and higher priced properties have not performed as well in the current market as smaller and lower priced properties. There are several reasons for this including: over supply of more expensive homes, the difficulty to obtain larger loans, and tax incentives for first time home buyers.
If the proportion of large properties that sold change in relation to the proportion to smaller properties that sold, then the RMLS statistic will again be skewed without a change in value. The following is a graph of average square footage sold over the two year time period in Area 145:
This graph shows that the proportions of larger and smaller properties have shifted significantly during the time of the RMLS reported Percent Change statistic. Over two years, the average square footage of sales shifted from 1,971 up to 2,445 and back to 1,994, a difference of more than 20% (the median shifted more than 30%). With this much swing in the size of properties being sold, one should expect a swing in the average price Percent Change statistic without a change in the market. This means the RMLS would report a decline in value when nothing actually happened.
A more reliable way to measure price trends, with less skewing from changes in house size, is to use price per square foot. The following is the same RMLS Area 145 Percent Change data trended using average sales price per square foot:
This data supports my theory from the Average Sales Price chart that no price drop occurred in 2010. The drop in Average Sales Price is likely skewing of data. This square footage data shows that an appraiser using the RMLS Percent Change of negative 9.5% to formulate time adjustments would be making a serious error. This is an error that we can’t afford to make in a slow market when comparable sales are often six months or older since the contract date. Even though average price per square foot is a more consistent and reliable trend, using it to support time adjustments could also be making a serious error.
As with square footage, there are other factors within RMLS Area 145 that could skew data including: submarkets, quality of properties being sold, condition or age, number of bank owned properties, etc… If the proportion of any of these factors changes in the market, the average sales price reported by the RMLS Percent Change can shift without any actual change in value.
One correct way to support time adjustments is thru using only statistics that come from sales of similar properties. Fannie Mae has just recently decided market statistics are important and introduced a new report form that includes some of the many statistics that appraisers should already be considering in the valuation process. When appraising, I chart only properties that are similar to the subject over two years and look at trends including, but not limited to: median list price, median sales price, sales price per square foot, inventory, market time, and REO activity. I’m a visual person, so it helps have some scatter charts or line graphs showing what is happening. It also helps my clients understand my conclusions. The two year time period allows for seasonal swings to be identified. The range or dispersion of the data tells how statistically reliable the numbers are. Using all of these statistics together with understanding the economy, market cycles, paired sales, interviews of agents, investors, buyers, and sellers produce a more reliable indication of the real market direction. Appraisal is not a science, it is an art. However, I believe that scientific data must be applied wherever possible to increase reliability and give context and support to a value opinion.
My hope is that the management at the RMLS will read this and increase the amount of data released in the Area Report in the following ways:
1. Each Area should have subcategories that track statistics for different residential property types, sizes, ages, lot sizes, etc… Maybe make the reports interactive.
2. Within the subcategories, median statistics and price per square foot should be reported wherever possible. These statistics are less susceptible to skewing.
3. Provide graphical displays for individual areas and not just regional data. Maybe add statistical measures of correlation or relivance.
We live in the information age. These recommended enhancements to the RMLS Area Report would have fewer people, including the media and real estate professionals, quoting misleading information. The recommended information is difficult and costly for individuals with access to the RMLS to compile because the RMLS only allows downloads of up to 300 sales at a time. The RMLS already has the data, someone just needs to tell the computer to track and report it.
If you read all that and you’re still wondering, what is happening in the Gladstone, OR market thru June 2010? The following is a summary of what I found in Gladstone while appraising three properties (I’ll try to keep it short):
· The 1,200 square foot house in Gladstone, Oregon market data shows that median sales price, median list price, and price per square foot only declined thru the end of 2008, were stable in 2009, and may have been increasing slowly in the first half of 2010. The bad news is that in the past three months, sales have declined and listings have increased.
· The Duplex in Gladstone, Oregon Duplex market data shows median list price, median sales price, and price per square foot have been declining consistently since the second quarter of 2009 (down roughly 14% in the prior year). The good news is that inventory has suddenly dropped and the all the REO properties have sold or are pending. Since REOs typically sell for less and depress the market, especially with vacant investment properties, median duplex prices are likely to improve (maybe sharply) in Gladstone this summer.
· The 2,900 square foot house in Gladstone, Oregon market data shows prices have been stable for the past year. This sample had the fewest number of sales and largest range in data, making the market condition estimates for this property less reliable than the other two Gladstone properties. However, inventory has been slowly declining for the past year in this category, supporting other data that suggests price stability.
Although my market analysis is more targeted to specific properties than some published data like the RMLS Market Action, use this information at your own risk. This analysis has been watered down from the version included in my appraisal reports and lacks the context and other supporting figures that those reports supply.
If you think you received a less than reliable appraisal from another company or have an example of another appraiser incorrectly applying statistics, contact us. We may be able to review it. Remember, appraisals can be challenged. If you found this information interesting or useful, please subscribe to my blog. If you need market data that is useful for a specific purpose without the cost of a full appraisal, contact us, we may be able to assist you. Thanks for reading.