Unfortunately, those cheery views all turned out to be incorrect. Why? It turns out that home purchases are very seasonal. There is a very well- defined pattern of sales from month to month over the course of a year. Reporting sales changes from January to February was actually measuring these seasonal differences. It was not, unfortunately, showing actual improvements in the housing sector. Of course, you would never report retail sales from December to January this way; otherwise each and every year, we would have headlines screaming RETAIL SALES FALL 65%. We know this to be the case because of two reasons. First, we can eliminate the seasonal variations by measuring the year-over-year sales changes. Second, we can look at the average monthly non-seasonally-adjusted sales data for the past few years to discern the basic calendar pattern. How did existing-home sales do in February based on the year-over-year data? Not particularly well: Single-family home sales in February 2008 were 23.8% below February 2007 levels. That doesn't sound very much like signs of stability of strength to me. The national median sales price was also a big surprise, dropping 8.2%. Those two data points reflect a very unhealthy housing market. However, rather than obsess over one month's reading, why don't we look back at the average monthly sales data, non-seasonally-adjusted, for each month. When we remove the seasonality factor, we learn quite a bit about home sales. Not surprisingly, it turns out January is the slowest month of the year when it comes to home sales. That intuitively makes sense, as most people are otherwise engaged in December -- they are busy with holidays and vacations. Many fewer people are shopping for homes then, so we get fewer contracts signed in January.
As the chart above shows, once we get past January, the lowest sales month of the year, its pretty much all good for existing-home sales through June. A pullback in July, a bounce back up in August, and that's the best first part of the year. Why? Well, consider how many families want to minimize the disruption to their kids' education during a move. It seems most families with school-age children want to be in their new homes before the new school year starts in September. This makes sense in terms of both the overall pattern of sales, and the flurry of closings in August. For a comparison, here is what the percentage change from month to month looks like using non-seasonally-adjusted data.
The big dropoff in September reflects the start of the school year. With so few buyers looking in December, its no surprise that January is the weakest month for sales. Finally, let's look at one last chart: Here is the non-seasonally-adjusted data for the past few years. Note again that February is better than January. March, April, May and June all see improvements also. While the numbers may change, slipping in each of the past four years, the seasonality of sales remains fairly consistent.
Source: Calculated Risk, NAR
Understanding this will help you contextualize the next few months of existing-home sales -- both the seasonal gains in the first half of the year, and the seasonal retreat in the latter months. But for the best insight into housing market, watch the year-over-year data.