Friday, March 18, 2011

Zillow ZHVI vs S&P/Case-Shiller HPI

Here is a comparison of the Zillow Home Value Index vs. the S&P/Case-Shiller National Home Price Index since February 1996. Zillow is blue; Case-Shiller is red.

The S&P/Case-Shiller index is a constant-quality index. Zillow's ZHVI is not. In order to compare apples to apples, I had to convert the ZHVI into a constant-quality index. I did so by graphing the ZHVI value per square foot. This way, the index is not distorted by changing home sizes.

Click the image to see a full-size version.

United States Home Values
In this graph, I think the ZHVI does a better job of showing where prices are headed.

Why the discrepancy between Zillow and Case-Shiller? I think the Case-Shiller index is distorted by foreclosures and potential home sellers keeping their homes off the market. The ZHVI ignores foreclosures. A given home is worth a lower sales price in foreclosure than in a regular sale, because the buyer takes a greater risk in a foreclosure sale and because the seller is a motivated seller. This means a foreclosure sales price is not a good estimate of a home's true value. The ZHVI also estimates home values not sales prices. This way, the index is not distorted by people who would like to sell their home but choose not to because of the declining market.

My U.S. housing bubble graphs use Freddie Mac's CMHPI for the 1970-1974 period, the FHFA HPI for the 1975-1986 period, and the S&P/Case-Shiller national HPI from 1987 to present. I am considering replacing the 1996-present period with the ZHVI. The goal is to use the most accurate data available for any given period.

Note: The numbers on the graph's y-axis are just index values, not dollar values.

9 comments:

  1. As A realestate investor in my own back yard town hse market valune 2007 was 149.000.0 before
    the bubble bursted.  I have A contract at 55,000.00. with futher decline on the way

    ReplyDelete
  2. ZILLOW caused the housing bubble. I wanted to kill myself after I saw my house plummet down by 100,000. I hope the CEO of Zillow kills himself.

    ReplyDelete
  3. I have always condemned Zillow because they don't know the difference between a 2000 sq/ft home on a mountain top, in the inner city or waterfront. It places both buyers and sellers at a disadvantage.

    ReplyDelete
  4. Zillow didn't cause the bubble. People did, out of greed and easy money from banks that were handing out mortgages like it was candy to anyone who asked for it, e.g., a maid owned 3 houses!!! and then later complained that she could not pay the mortgage.  Come on!!!  Think with you head people, not with your bottom.

    ReplyDelete
  5. BS.   Zillow confused my neighborhood (waterfront) with a new development in foreclosure with hundreds of homes, in a completly different place (low rent) and based on this error, their Zestimates for this area dropped 40% over night.   I didn't sell, but some had to (at the low price) which then became a self fulfilling prophecy, as Zillow incorporated those sales into the NEW prices.  They then used that negatively biased  (and wrong) information to bias their algorithms for the town, then the city, then the state.  This spiral took things in a stable, waterfront community with low turn over, low crime, good schools, in an area of low unemployment, and a growing population etc.  down 60% in value on average.

    In doing a little research, I saw this same thing happening ALL over the area, so I searched more.  I then saw it all over the state, and the country.  Zillow was new, and people didn't know what to do, but still, thousands of people complained about it very specifically, and Zillow just buried it.   It is interesting to see now that they are systematically changing their "past" data, including deleting their past predictions.  This looks to me like they are covering their tracks.

    Zillow should be sued for negligence for the way they haphazardly used data (wrongly) in many cases, that caused a downward spiral in an entire  market that wasn't ready for the misinformation...

    ReplyDelete
  6. I agree, and this isn't just a rant.  There is science behind "how" this happened.  Zillow took data and used it for something that it wasn't intended, and didn't "clean" the data.  Much of the data was wrong, and they actually added errors by putting houses in the wrong place etc. which made some estimates VERY low, which in turn (along withj a slight negative bias) drug the market estimates down.  Most people (that could)  just held on, but some couldn't because of job loss, divorce, death etc. and had to sell at the low estimates.  This started a downward spiral, because Zillow used that sale data to affect other estimates, bringing those down further, and as time goes on, a percentage of owners MUST sell, again at whatever price, making Zillow's low estimates a self fulfilling prophecy.   

    I have even noticed that Zillow has changed their graphs of housing values for the "past" where they were wrong, and now try to cover their tracks, including deleting data all together.  They keep up this Charade in order to take power over the market, and make real estate people desperate enough to advertise with them.  You can see them trying to do this (again) now with property managers and real estate investors, with their rent estimates, that are just as much all over the board as their Zestimates

    Zillow should be sued for their "Zestimates", for erroneously using wrong data, and in so doing manipulating markets and being a major driver in the historic housing decline, causing huge losses to average Americans, ruining their lives, and harming the US economy to the tune of Trillions.   

    ReplyDelete
  7. 3240 E Pinchot in Phoenix is a VACANT LOT that zillow estimates at $138,000.  It was listed last year at 40,000 for six months and didn't sell, so zillow is off by a factor of more than 3.  And it's not due to outdated information about the number of bathrooms.

    ReplyDelete
  8. Wait, why are people treating Zestimates as truth?

    ReplyDelete
  9. I make the point on http://homepriceevaluator2.blogspot.com/ that we should be looking at median house price per median income to judge affordability. I will update my findings to make a graph of historical data. As a rough estimate, I view today's houses at more than 2X more costly than in 1960. When using a comparison like this one does not have to be concerned with inflation as whatever the claimed inflation rate is, the measure of affordability will have to relate to the income at that time. There will always be problems associated with any summary statistic though and for this would be the size of homes, but could be offset by a smaller plot area, the median income as far as all sources of income and not taking from only those considered as working, and some others. I will make a case that we shouldn't concern ourselves with mortgage interest rates while within some normal range as this can be a factor of loose/tight money supply and it becomes moot if you are an all cash buyer as some responsible persons aim to be (discounting all the financial muck that you can make more in the stock market or elsewhere and not pay off the house, and all this mortgage interest deduction which I say should never have been part of the IRS boondoggle as it benefits the wealthy far more - say if you do not work but saves lots of money- because you were responsible you get no kickback!).

    ReplyDelete