Housing data crybabies and deceivers

With a name of a publication like the one you see at the top of this blog, if there are any biases by the writer (which surely there are), you probably have a good idea what they are. Not so with outfits like MarketWatch where columnists can write whatever they want while keeping their true motives unknown to readers.

Either that, or they just don't understand the data.
Such was the case in yesterday's report by Chris Pummer about how recently reported home price declines of unprecedented magnitude should not be trusted.

Writers should be careful when prying into data reporting, especially when they side with Lawrence Yun and the National Association of Realtors.

Chris writes:

Commonly cited measures of U.S. home prices are overstating the degree to which the vast majority of Americans' home values have declined in the last year, producers of two of the most widely tracked indexes acknowledged this week.

Yes, all statistics are flawed - some more so than others. Get used to it.

More importantly, where did that "vast majority of Americans' home values" line come from? That's not what the NAR or S&P said about their home price indexes.

There are two arguments here, both of which are valid, neither of which makes the condition of the nation's housing market demonstrably better than reported in the mainstream media.

First is the "misleading median" that tends to be pulled up or down depending upon the mix of high-priced and low-priced homes being sold:

"If there are a lot more homes sold on the low end and fewer on the high end, the median price is bound to drop dramatically," NAR Chief Economist Lawrence Yun said. "In normal times, a median price would reflect typical homeowner equity changes, but these are not normal times. The jumbo (mortgage) market is frozen and the buying activity is more concentrated in lower-value homes."

Yes, that's true - the only problem is, this works both ways and unless Chris was one of the very few writers who complained last year around this time, then he should pipe down today.

Early in 2007, when the subprime problems were gathering steam and lender's were cutting back on making loans to shady borrowers, the sales mix shifted to the high end as jumbo loans were still considered to be "safe". This pushed the "median" higher than it would otherwise have been.

Once the credit crunch hit in August and jumbo lenders started to make up for lost time, realizing that they too might not get paid back after home prices started to fall, a sharp pullback in jumbo loans has skewed prices downward.

If you didn't complain about the first, you can't complain about the second.

The next arguments for home price declines being overstated involve the limited coverage of the Case-Shiller Home Price Index (yes, this is true - that's why they call it a 20-city index - it only covers 20 big cities) and one of the loopiest bits of reasoning I've read lately.

The S&P/Case-Shiller index, which Tuesday posted a 12.7% decline for February, is skewed for two reasons of its own -- it tracks just 20 major markets, many among the hardest hit, and its "repeat sales" survey by design pulls in individual homes both bought and sold in the last few years. Many of those are now being dumped by distressed homeowners and investors who bought at peak market prices and face higher mortgage-rate adjustments.

In other words, ignore how prices got to such lofty levels while the housing bubble was inflating and remember that a home hasn't declined in value until it is sold.

It's only the distressed properties that are being sold at much lower prices - forget about the impact that these prices have on determining the value of other homes. As soon as this multi-year wave of foreclosures passes, things will be back to normal.

How dumb!

But Chris' biggest mistake when complaining about other statistics being misleading (and the real motivation for going off on this rant this morning) comes when he cites NAR data to counter what is being reported in the media:

The glaring discrepancy in this case is that 17 of the 20 metro areas posted record annual declines, and yet 78% of the 330 metropolitan regions that NAR tracks reported price increases in the latest period -- and that despite the acknowledged downward bias in current price readings.

The most recent Metropolitan Median Price data from the NAR which included data through the fourth quarter of last year is available for anyone to see at the NAR website - while overall prices declined during the fourth quarter, 49 percent of the metroplitan areas showed price increases. Housing hotspots such as Yakima, Washington and Bismark, North Dakota top the list of areas with price increases.

Talk about misleading statistics!

The data cited in the MarketWatch story - 78% of metro areas showed increases - is actually from the third quarter report, data that was used liberally by the NAR to refute misleading reports of home price declines up until the fourth quarter report came out.

This most recent data is up to seven months old and the data cited in the MarketWatch report is an astounding 10 months old, going all the way back to July of 2007 before the credit crisis began.

Appropriately, the piece concludes with more wisdom from Lawrence Yun, who really does seem to be choosing his words much more carefully than his predecessor:

"The only way to tell what your own home is really worth is to look at local-market conditions, do Internet research and utilize professionals (such as licensed appraisers) to help determine the value of your home."

OK.

Nothing is selling. They're dropping the prices of bank-owned properties faster than they were a couple months ago and the banks can't seem to keep up with the mounting number of properties that they are collecting from buyers who either can't or don't want to pay their mortgage anymore.

There are a huge number of sellers out there with prices that are far removed from what they could reasonably expect to get today, no thanks to the claptrap coming from people like Chris Pummer.