Property price expectation weakens as property sales rise - why?

At times I have to confess I get a bit lost in analytics and data analysis, trawling through the reams of property data available - I love to seek out trends and then try and interpret the correlation. I have come across just such an interesting correlation and am curious to postulate some rationale behind the correlation and see what others think.

The correlation is that there is an inverse relationship between property sales volumes and the alignment of the asking price of property to the eventual sales price. That is to say that as sales volumes rise then the differential between asking price and selling price lessens whereas, when sales volumes fall, the gap widens. Or put another way, sellers seem have a more realistic price expectation in a rising market! - not what I might have thought.

I was prompted to this correlation when examining the current trend of asking price to sales prices over the past fiv years. The source data I have looked at is the Stratified Price Index from the Real Estate Institute and the Asking price from the Realestate.co.nz NZ Property Report which uses a 80% Truncated mean price. I have chosen to examine the Auckland market specifically as it has witnessed sizeable movements in both volume sales and prices over the past 5 years.

In the past year the asking price for Auckland properties coming on to the market has actually fallen below the sales price or, put another way, the pace of sales prices has outstripped asking price. This was the beginning of the data trend analysis. I next added the data of property sales over the period and tracked this against the variance of sales price to asking price.

The red line represents sales tracked on the left hand axis whilst the grey line tracked on the right hand axis shows the percentage variance of sales price to asking price with the parity level marked at the 0% horizontal line. A very clear correlation is seen albeit the monthly fluctuations make the chart a trifle messy.

Applying a 12 month moving average to the data sets though provides a clearer picture of the correlation.

So having established this correlation, the question is how can this best be explained? Here are some thoughts:

1. Property sales are a lead indicator of price movements and, as can be seen, there is a lag in the trend correlation of anywhere from three to nine months. If asking prices are set based on current market demand and sales prices follow by a lag, then the gap between asking price and sales price will narrow as sales consistently grow.

2. Asking prices are set more by agents than by vendors. As sales pick up, agents are motivated to sell and therefore set more realistic expectations of asking prices especially at the early stage of sales growth. On the downside, as sales fall, agents hold out for a more optimistic selling price than the market would indicate.

3. In Auckland, as we are so often told, a majority of property is marketed without a published price and often for sale by auction. The asking price is created by the website search price, which is not public. Therefore it could follow that as auctions become more popular (as sales growth takes off), the search price is set at the bottom end of price range expectation to attract attention, which leads to this narrowing in the gap. Conversely, the fall in sales drives less auctions and more realistic search range pricing.

4. The widening of the gap between asking price and selling price is possibly explained by the fact that expected asking prices still reflect an optimism by agents to see price continue to rise when the reversal of sales growth actually seeing property prices fall.

Whatever the explanation, I think it is interesting to see just how close asking prices have come to the selling prices. The property sales have started to tail off and therefore the acid test will be if the gap between asking price and sales price widens driven by continuing growth in asking price expectation.