Non-governmental Statistics

After having looked at some of the stats from both the CRIS MLS and the Office of Federal Housing Enterprise Oversight ( http://www.ofheo.gov/media/pdf/1q07hpi.pdf ) I had to laugh. "U.S. House Price Appreciation Rate Remains Slow, But Positive", says the headline. Then come the stats, and a nice bunch of graphs which show prices nose-diving since the beginning of 2006.

The OFHEO table shows a +0.84% change in house prices in Ohio for the one year period ending March 31, 2007. In contrast, the MLS data published in Ohio Realtor in January 2007 ("Ohio housing market posts second best-ever mark in 2006") showed a statewide decline of 2.1% in sale price in the period January - December 2006, and in April 2007 the published data for January - March 2007 showed a further 1.5% decline.

Therefore, it simply makes sense not to rely on headlines, especially government publications, unless you want to be fleeced. The best way to handle the situation is to crank out your own analysis, and a good spreadsheet is a valuable tool -- if you know how to feed it, use it, and understand the implications of its output.

The feeding of the spreadsheet is somewhat problematic. I am still looking for ways to feed the output of an MLS statistical report into a database, other than typing each field individually. I just set up a report template for Congress Township in Wayne County, and the public records data for the market trend analysis was drawn from PACE books. For what I needed to do immediately (the assignment was a Land Appraisal) I simply typed in the parcel number, the closing date, and the sales price, then let the spreadsheet handle finding the high, low, mean, median, and mode for each year.

I quickly realized that the data I had input was insufficient, considering that multiple transfers showed up and it became apparent that with long-term data it would be possible to track individual homes to measure possible appreciation in raw sales price. Coupled with the fact that many of the transfers that occur involve multiple parcels, it would be possible to refine site value extraction using that data. Using any MLS data, which could describe changes in homes between sales, it would also be possible to measure true appreciation as opposed to a rise in sales price caused simply by improvements to the home.

Why go to this trouble? I have recently been contacted and engaged by the U.S. District Court to perform appraisals on properties involved in Federal litigation. Three separate appraisals are required for each property, by different appraisers. I have no doubt that the results of the appraisals will vary, and depending on who is engaged for the work, the results may vary widely. I need to be ready to defend my work in the face of statistics published by the government.

There is also the specter of a tax increase due to the required sextennial reappraisal of all parcels in Ohio. Unquestionably, the value of many homes is higher today than in 2004, when the last mid-term adjustment took place. The rate of increase, however, may be open to dispute, since the Auditors statewide will be feeding government generated data into their AVMs. As with all such reappraisals, there are many people who contest the change in the appraised value of their homes, and hire an appraiser to challenge the Auditor's office. I need to be ready for that, and if a challenge occurs, be ready to defend my work.

People keep asking me if I enjoy my work. I think I'm having fun.