The NAR reported a decline of 3.2% in existing home sales in July. This was the seasonally adjusted fudgepack number. It prompted the Dow Jones flagship Wall Street Journal to trumpet:

U.S. Existing Home Sales Fall for First Time Since February

I prefer to look at the actual data, using the annual rate of change as the basis for comparison. To see whether momentum is slowing or growing, we can compare the annual growth rate of the current month to recent past months. Actual sales in July fell by 6.7% year to year.

So far, so good.

However, there were calender factors involved. July 2015 had 22 business days available for closings versus only 20 in July of this year. While some of those closings would still have been squeezed into last month toward the end of the month, some adjustment to the number would be necessary. On that basis it’s likely that the number of closings on a rolling 4 week or 31 day basis would not have been materially different this year versus last year. Compared with a year to year change of +1.7% in June, it appears that the growth in sales volume has merely stalled, not plunged.

That is supported by the June year to year contract data, which the NAR calls “pending home sales.” Contracts were flat in June. That would normally mean flat growth in closings in the ensuing month.

There’s some evidence that the number of contracts failing to close within 2 months of the initial contract month may be starting to increase. However, it’s too early to call either this or the volume stall a significant change in market conditions.

Low inventory is still a problem. It is partly caused by Fed low interest rate policies which discourage older homeowners from liquidating their properties when they retire. Normally rapidly rising prices would cause an increase in the supply of homes on the market. In this case, they have not.

Artificial mortgage rates also stimulate demand that would not exist were it not for the suppression of mortgage payments allowing more buyers to qualify. As a result, housing prices continue to inflate. The national median price is now 6% above bubble peak levels. There are those who argue that that’s not a bubble. If it was a bubble in 2003-2007, why not today? Because they want you to forget history.

Meanwhile, median household income has lagged house price inflation since 2011, causing affordability to be a growing problem. Prices have risen by 43% over that period. Median Household Income has only risen by 11%.

Obviously, this can’t continue for long without causing the market to dry up. Volume growth before the fall. Sales volume fell sharply between 2005 and 2006. While prices were modestly higher in 2006 and 2007, selling your home in 2007 was like winning the lottery. Demand had completely dried up. A persistent decline in volume over several months today would be one of the first signs that the market had peaked. So far, it has only gone flat.

House price inflation accelerated to an annual rate of +5.3% for July closings. This indicates slight acceleration from May and June when price increases were 4.4% and 4.8%.

This data is from the MLS and is far more timely and accurate than Case Shiller. That index relies on government data on deed recordings which lags closings by 4-8 weeks. In addition, Case Shiller uses a 3 month average price which increases the lag. And Case Shiller only includes repeat sales, which creates a selection bias to older, smaller homes, which are likely have depreciated physically, if not functionally. Case Shiller data is thus severely lagged and understated. It is not a good indicator of current market conditions. The NAR data is more timely, and the not seasonally adjusted numbers are actual compilations of all MLS data from throughout the US.

As a former professional commercial real estate appraiser, real estate broker, and mortgage broker with 20 years experience in that industry, I consider this to be good quality data. The spin that NAR PR people put on it is not, but it’s easy enough to look at the raw data, draw a graph, and see what’s going on.

For now, we’re in a status quo trend where central bank policies artificially suppress mortgage rates. That creates artificial demand and limits the supply of homes for sale. That in turn results in rapid price inflation which is racing ahead of household income gains. This is not a process that can continue indefinitely. A 43% gain in house prices over 5 years when incomes are only up 11%, will at some point lead to the next housing crash. We’re not there yet, but after 5 years of artificially stimulated recovery, the risks are growing.