< All Articles & Research Papers

Real Estate Market Statistics: User Beware!

Jonathan J. Miller
April 8, 2001, 3 pages

Along with the changes in the Manhattan real estate market over the past year, there has been an increased appetite for related information. With the advent of desktop publishing, most real estate firms or organizations have a real estate market study, statistics and web site to share information with the public. Our firm prepares several such market reports for the brokerage firm Douglas Elliman. The purpose of this discussion is to suggest that users of real estate market information find ways to look deeper into what is before them and test its validity. Not to simply rely on information that is “spoon-fed.” Most often, market information is readily available but is not presented in a reliable manner.

“A billion here, a billion there, sooner or later it adds up to real money.” – commented Everett Dirksen, Illinois Senator 1951-69 concerned with President Truman’s domestic spending on the Great Society. Our study for the year 2000 contained roughly $8 billion worth of transaction information but it is simply not enough for the public appetite. The same phenomenon has been true with the financial markets. Historically, the financial markets change much more rapidly than the real estate markets and now there are tickers, web sites, etc. dedicated to the dissemination of information. Real estate is in its infancy in this regard.

Lies, damn lies, and statistics

Benjamin Disraeli, the prime minister of the British Empire from 1874-1880, was reported by Mark Twain to have said: “There are three kinds of lies; lies, damned lies and statistics” Viewing real estate market statistics is more than simply a matter of looking at one chart and seeing if the price went up or down. The media tends to report the overall market in singular terms and users tend to accept it. Everyone is looking for that one market signal. Extreme statements are usually what make the evening news or tabloid headlines. “Apartment prices skyrocket!” is much more interesting than “modest appreciation occurred in all market segments.” To make matters worse, users over-interpret what is actually being said. Just recently, one of our market statistics was that “the overall median sales price increased 50% from the same period last year.” This doesn’t mean that it represents market conditions in the segment you are interested in. In this particular case, the 50% rise was more attributable to the shift to larger apartments since the individual size categories only (the term “only” is probably a bit cavalier) increased about 25% to 30%. Last spring we reported that double digit price increases of the past year seemed to be abating. The rate of appreciation was easing but the market was still rising. I found it interesting that many market participants seemed to interpret this news as “Its over. Prices are falling.” when the discussion of falling prices was nowhere to be found in the article and to this date we have seen no such evidence. I think the way to view real estate market statistics is to:

Be skeptical. – Use them to search for a trend, not a number. – Look at as many statistics as you can find. – Consider different types of statistics, not just sale prices. – Focus on a segment that drives your interest in the market. – Consider the source: Is the content provider motivated to sway your opinion?

And finally,

  • YOU CAN’T PLACE ALL YOUR FAITH IN ONE MARKET INDICATOR. There are simply too many indicators available to you that can not be ignored.

Median sales price Some observers prefer the median sales price as an indicator over the average sales price because the highs and lows are removed. The National Association of Realtors is widely referenced using this statistic. However, in some real estate markets, the median sales price could indicate that the market is still rising when it may have leveled off or begun to decline. This is how it works: In a market with 100 sales, the 50th sale is the mid-point or median. All sales above the median are more expensive than the median and all sales below are less expensive. However, as the market begins to deteriorate, quite often the entry-level market erodes first by recording fewer sales. Now those 50 sales below the median become, say, 25 sales and the median sale price is driven upward. The statistics therefore show a rising market with lower volume. Although that is a fair description of the current market, all our market indicators in each size segment still show strong evidence of a severe demand over supply imbalance. There is no near term solution to the lack of inventory that has characterized the market over the past several years. The point here is that median sales price should be used in tandem with average sales price as well as other statistics that are available.

Market study

A real estate market study, or at least our studies, are simply an overview and can be a helpful tool in estimating the price of a given property. However, it is not designed to be the sole determinant of price. It is more useful in establishing a general trend of a given market segment. There are dozens of amenities that are blended into these statistics: pre-war, post-war, doorman, nondoorman, loft, non-loft, view, non-view, renovated, non-renovated, etc. that may make the property unique. Getting that specific would necessitate either a broker price opinion report or a real estate appraisal report, but that is for another discussion.

Wonder of the data

I find many of the graphics; charts, tables, graphs on the real estate market hard to read or even worse, misleading. One of my favorite all-time experts on this subject is Edward R. Tufte, Professor of Political Science, Statistics and Computer Science at Yale University, author of Envisioning Information and several other notable publications. He believes that bad graphics “lie by distortion, obfuscate by omission and confuse by decoration.” If we spend so much time trying to figure out the meaning of a graphic, we miss out on the “wonder of the data.” His position is that any chart or graph without the source or critical item like the size of the data set “is a lie.” I absolutely agree. A sore point for me is reviewing statistics that are based on a small data set, or the actual size of the data set was not disclosed, or the reliability of the data is questionable. It has been reported that a Manhattan market study done by another firm many years ago was based on only 10 sales. This fact was not disclosed at the time of publication. Was that done for the public benefit or was it self-serving?


One of my favorite quotes is: “Professionals are people who can do their best at a time when they do not particularly feel like it.” – Alistair Cooke. Presenting clear and accurate information is hard work. The greatest challenge for our company, Miller Samuel, has been to present a large and thorough cross-section of data, analysis, graphs and charts as clearly and accurately as possible. It is a fun but never-ending effort and we have benefited greatly from user feedback. Our market reports and maintenance of our web site – MillerSamuel.com – have helped keep our firm focused on pertinent market conditions when preparing our appraisal reports. We know what we are presenting to the user – that is the easy part. The hard part is communicating an accurate and relevant snapshot of the market. We want the user to benefit from our resources and to give us their trust for the next time they need information.

Copyright 2001 Miller Samuel Inc. All World Wide Rights Reserved.

Download Now

Get Weekly Insights and Research

Housing Notes by Jonathan Miller

Receive Jonathan Miller's 'Housing Notes' and get regular market insights, the market report series for Douglas Elliman Real Estate as well as interviews, columns, blog posts and other content.

Follow Jonathan on Twitter

#Housing analyst, #realestate, #appraiser, podcaster/blogger, non-economist, Miller Samuel CEO, family man, maker of snow and lobster fisherman (order varies)
NYC CT Hamptons DC Miami LA Aspen
Joined October 2007