Two housing market indicators released over the last two days suggest that the market is making a seasonal peak at a weaker level than last year, but that the rate of decline is slowing.
Mortgage purchase applications have been slipping in June following a small decline in the week ended June 17. They remain below the long term trendline and right at the 52-week moving average. There's little sign of recovery here, but demand has at least stopped weakening over the past year. On the basis of the seasonally adjusted data, applications are now up 4.4% versus this date last year but they are still down 36.2% since the second homebuyers' tax credit contract signing deadline of April 2010. They are down 64.7% from the bubble high.
The Mortgage Bankers Association Mortgage Purchase Index is the closest thing we have to a real time housing market activity index. Unfortunately my chart only shows the seasonally adjusted data, but at least that gives an idea of where the trend is headed over time.
Click to enlarge
Bloomberg has the actual, not seasonally adjusted data chart here.Those of you who have a Bloomberg terminal can do some interesting analytics with this data. It shows applications rising 4.8% in the past year but down 38.3% since the tax credit scam peak. Notably, the seasonal peak has typically been in May or June. This year, it appears to have come in April. Purchase applications in the week ended June 17 were down 15% since the April peak. That's going in the opposite direction of what it typically should be at this time of year, but it's consistent with the trend of the past three years.
Click to enlarge
The NAR reported on Tuesday that cash sales accounted for 30% of all transactions, up from 25% in May of 2010. Prior to 2010 the NAR did not routinely report all cash purchases since they were not a significant part of the market. With so many cash sales now, the mortgage purchase applications index is probably understating market activity, but cash sales are often vulture investor purchases that do nothing to solve the supply demand imbalance. At the same time both the high percentage of cash sales and the apparent slowing of the downtrend in mortgage purchase applications suggests that the market has reached a demand floor. That could change if the number of total employed persons begins to decline again. I will continue to monitor that data monthly in our Professional Edition housing updates.
The NAR's existing home sales prices in May continued their rebound from the low in February. Because the NAR does not use a moving average like Case Shiller, its data is a bit more timely, but it also reflects seasonal bumps. A seasonal bump in prices from the winter low to a June or July high is normal. The question is whether that rally makes a lower high or higher high, or whether it shows a change in the momentum of the decline. The median sale price rose from $161,100 (revised up) in April to $166,500 in May. That is still 4.6% below last May's level. The gain appears to be slightly less robust than the seasonal uptick in prior years. Prices usually peak in June, plateau for a month, and then head down. Unless there's a big price surge in June, the odds are that prices will make lower lows in the winter.
The best real time pricing data comes from Housingtracker.net, which correctly tipped us off in real time that prices began rising in March and continued to do so through May. The early returns for June, through two days ago show a mixed bag. Of the 10 big metros in the Case Shiller 10 cities index, four had no change in price level over the past month, four had gains ranging from 1.1% to 3.3%, and two had losses of 0.8% and 2.3%. Prior to this, usually all but one metro has shown a gain each month since March. This suggests that the seasonal price peak is at hand. I will cover this in depth in our next Professional Edition housing update.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.