In Australia, flawed housing data draws some concerns over their monetary policy thinking [The Australian].  The Reserve Bank of Australia (RBA is the US equivalent of the Federal Reserve Bank) has been increasingly vocal over the past few years over the quality of housing statistics. “Most of the available price information from real estate agents, banks, the various organisations that monitor house prices and the Australian Bureau of Statistics is flawed to some degree.”
Aside: the housing peak in Sydney was in 2003 and the market has fallen 7% in nominal terms over the past 18 months.
The concerns raised by the RBA are the timeliness of the information, changes in the mix of housing being sold (ie, large, small, etc.) and quality and size of the housing being constructed.
Same issues in the US
We have similar issues with the housing data available here. Census data is delayed. We rely on NAR for a large portion of the statistics but they are an industry trade group – by definition has a bias, and the data is generally lumped together not reflective of changes in housing mix.
The Federal Reserve does not appear to be displeased with the US data available to the public. Are they looking at something else?
Saw a great quote published in a recent issue of The Economist in an article on the accuracy of scientific research papers that may be appropriate when applied here:
Theodore Sturgeon, An American science-fiction writer, once observed that “95% of everything is crap”. John Ioannidis, a Greek epidemioligist, would not go that far. His benchmark is 50%.