In Depth Analysis: CalculatedRisk Newsletter on Real Estate (Ad Free) Read it here.

Friday, July 01, 2011

Lawler: Do Housing “Experts” Actually Look At All Available Data?

by Calculated Risk on 7/01/2011 08:24:00 PM

CR Note: I think this is pretty funny ...

Economist Tom Lawler writes: “Consensus” Home Price Expectations from MacroMarkets; Do the “Experts” Actually Look At All Available Data?

MacroMarkets reported two weeks ago that the average forecast for the growth rate in the S&P/Case-Shiller national home price from Q4/2010 to Q4/2011 of the 108 “economists, real estate experts, investment and market strategists” in its June survey was -3.52%, down significantly from -1.38% in March and -0.17% in December. The median forecast called for a 3.1% decline. In addition, the average forecast for the SPCS HPI from Q4/2011 to Q1/2012 fell to +0.46% in the June survey from 1.26% in March and 3.07% in June.

MacroMarkets Survey Results, Projections for the S&P/Case-Shiller "National" Home Price Index
% Change, Q4/2009 - Q4/2010% Change, Q4/2010 - Q4/2011% Change, Q4/2011 - Q4/2012

Obviously the “surprising” weakness in the SPCS HPI in the first three months of 2011 was the major catalyst for the substantial reduction in the average and median projection for 2011, and I’m guessing that some of the weak economy numbers over the last few months contributed to the drop in the 2012 projections as well. In the June survey participants did not have the April SPCS “composite-20” report yet, but other HPIs showing a bounce up in April were available – though I’m guessing many of the “expert” participants don’t look at all of the data available.

Indeed, the “abysmal” track record of the “consensus” forecast for 2010 highlights how many of these “experts” just don’t follow the markets and incoming data. While in June of last year the average projection called for a 1.36% drop in the SPCS HPI from Q4/09 to Q4/10, in November the “consensus” had improved to show a 0.54% decline. More shockingly, of the 95 survey participants who allowed their forecasts to be published (some don’t), an astounding 32 of them were still projecting a Q4/09 – Q4/10 increase in the SPCS HPI last November – which just seemed inane based on all of the data available at that time. In the November 2010 survey, 55 of the 95 participants with disclosed forecasts missed the “actual” 3.79% decline (Q4/09 – Q4/10) in the SPCS national HPI by over three percentage points – 54 were too high, and one was too low! And this was in November!!!!! (I was off by 0.79 percentage point, which I viewed as a disturbingly high miss, yet only 5 survey participants -- of which I am one -- were closer than me!)

CR note: This was from Tom Lawler. Obviously the "consensus" is suspect. We've seen this several times in recent weeks. As an example, this week Lawler correctly predicted a significant rebound in the Pending Home Sales index, and the "consensus" was for a 2% decline!

Back in May I pointed out the "consensus" for employment, auto sales, and the ISM manufacturing index all seemed "too high" (I was being polite) - and the misses were signficant. Just something to remember when looking at the consensus ...

ISM Manufacturing index increases in June
U.S. Light Vehicle Sales 11.45 million Annual Rate in June