For a long time I have had a disdain for the Case-Shiller Index that we have talked about here many times. I have always thought it took a 10 and 20 city sample and casually and flippantly interpreted and huge and complex market from those numbers. Case-Shiller IMO is about as meaningful for the individual’s home value as the DJIA is for an individual stock. CS is backwards looking and also captures the extremes (mass investor buying, foreclosure sales etc).
I first began thinking harder about this last year through empirical evidence. Homes in my area and of people I knew in other areas across the country just were not selling for the huge discounts I was hearing about. In fact many were seeing stable or rising prices that directly contradicted CS data. I came to the conclusion there were in fact had two housing markets, the distressed and non distressed. Because if that, any “index” that captured both into one and gave a general pricing level was fundamentally flawed.
In my search to find an index that more accurately reflected what was going on I came across the FHFA
FHFA is more meaningful as it captures not what is being done at the high and low end of the housing, but the massive middle. It is telling us what conditions are like if you are a unstressed homeowner selling your home or the ordinary buyer. It eliminates the price depressing effects of the investor in Detriot buying 25 lots for pennies on the dollar and the investor buying 15 Miami condos for 30% of the previous purchase price. That sale does not adequately represent the general price level of homes. The extremes skew the averages. When we were pushing through the huge mass of foreclosures, that artificially lowered the perceived housing pricing level
How does the FHFA HPI differ from the Case-Shiller® Home Price indexes?
Although both indexes employ the same fundamental repeat-valuations approach, there are a number of data and methodology differences. Among the dissimilarities:
- The S&P/Case-Shiller indexes only use purchase prices in index calibration, while the all-transactions HPI also includes refinance appraisals. FHFA’s purchase-only series is restricted to purchase prices, as are the S&P/Case-Shiller indexes.
- FHFA’s valuation data are derived from conforming, conventional mortgages provided by Fannie Mae and Freddie Mac. The S&P/Case-Shiller indexes use information obtained from county assessor and recorder offices.
- The S&P/Case-Shiller indexes are value-weighted, meaning that price trends for more expensive homes have greater influence on estimated price changes than other homes. FHFA’s index weights price trends equally for all properties.
- The geographic coverage of the indexes differs. The S&P/Case-Shiller National Home Price Index, for example, does not have valuation data from 13 states. FHFA’s U.S. index is calculated using data from all states.
The FHFA purchase-only index is based on more than 6 million repeat sales transactions, while the all-transactions index includes more than 45 million repeat transactions. Both indexes are based on data obtained from Fannie Mae and Freddie Mac for mortgages originated over the past 37 years.
FHFA analyzes the combined mortgage records of Fannie Mae and Freddie Mac, which form the nation’s largest database of conventional, conforming mortgage transactions. The conforming loan limit for mortgages purchased since the beginning of 2006 has been $417,000. Pursuant to the terms of various short-term Congressional initiatives, loan limits for mortgages originated between July 1, 2007 and Sept. 30, 2011 were as high as $729,750 in certain high-cost areas in the contiguous U.S. Mortgages originated after Sept. 30, 2011 were no longer subject to the terms of those initiatives and, under the formula established under the Housing and Economic Recovery Act of 2008, the “ceiling” limit for one-unit properties in the contiguous U.S. fell to $625,500.
How is calculated?:
The House Price Index is based on transactions involving conforming, conventional mortgages purchased or securitized by Fannie Mae or Freddie Mac. Only mortgage transactions on single-family properties are included. Conforming refers to a mortgage that both meets the underwriting guidelines of Fannie Mae or Freddie Mac and that does not exceed the conforming loan limit. For loans originated in the first nine months of 2011, the loan limit was set by Public Law 111-242. That law, in conjunction with prior legislation, provided for loan limits up to $729,750 for one-unit properties in certain high-cost areas in the contiguous United States. For loans originated and acquired by the Enterprises after September 30, 2011, the applicable loan limits are set under the Housing and Economic Recovery Act of 2008 (HERA). The HERA limits do not exceed $625,000 for one-unit homes in the contiguous United States.
Conventional mortgages are those that are neither insured nor guaranteed by the FHA, VA, or other federal government entities. Mortgages on properties financed by government-insured loans, such as FHA or VA mortgages, are excluded from the HPI, as are properties with mortgages whose principal amount exceeds the conforming loan limit. Mortgage transactions on condominiums, cooperatives, multi-unit properties, and planned unit developments are also excluded.
The way the index is calculated virtually eliminates the investor buying of REO’s (most, not all, but far more that the other indexes). These people are not getting conforming loans for these purchases. In fact in the coming months FHFA will have its data gathering refined to the point they will be releasing a “Distress-Free Index” that is currently being tested in Arizona. Mandatory reporting on this will eliminate REO and short sales from pricing data. This will allow for truer pricing. It will go national once refined:
Sales of bank-owned properties and short sales—collectively known as distressed sales—occur at significant discounts relative to other transactions. As has been discussed in previous FHFA publications, 1 price trends reflected in the FHFA HPI and other commonly-referenced real estate price metrics can be substantially influenced by such transactions. Fluctuations in the share of FHFA’s data sample comprised of such sales will affect measured price trends. For example, if an unusually large percentage of FHFA’s sample is comprised of distressed sales in a given quarter, the price change reported for the quarter, all else equal, will tend to show greater price weakness.
Some users of the FHFA HPI have expressed interest in having “distress-free” indexes estimated on data samples that exclude distressed transactions. Prices for properties sold in distress tend to be lower because of poorer property condition and stronger- than-usual seller motivation—factors that, for some purposes, might be appropriate to exclude. For example, when estimating home values and associated statistics such as the loan-to-value ratio for homes whose homeowners are not in financial distress, a distress-free measure might be more relevant.
Ok, great Todd, why does any of this matter and can you prove it?
Thanks to my buddy Fred, I can:
The red dot is the March # of 2.9% (Fred has not update their data set yet).
Couple of items. Notice the FHFA data did not spike as much in ’09-’10 due to the 1st time home buyers credit (was a tax credit for individuals and couples who purchase a new home after April 8, 2008, and before May 1, 2010). Why? It is a wider sample (National), not just 20 or 10 cities. Also, go back to