F reddie Mac reported last week that home price growth for the fourth quarter of last year increased significantly over the previous quarter, growing at an annualized rate
of 17.8% nationwide - a number not seen since 1976. However, analysts do not believe this figure accurately reflects the actual quarterly appreciation.
The GSE also reported in its quarterly Conventional Mortgage Home Price Index (CMHPI) that home values rose 8.4% from the fourth quarter of 2002 through the last quarter of 2003. This figure reflects an increase from the prior year (4Q01 to 4Q02) when the house price growth rate was only 7.6%.
This considerable rise is a result of fewer "revision" and "appraisal" errors. Revision errors happen when loans originated late in the quarter don't reach the GSEs in time for the quarterly Index release. Meanwhile, appraisal errors result from reused appraisals on streamlined refis. Both of these cause home prices to be undervalued.
The end of the refinancing boom has put a stop to these errors, finally providing a more accurate picture of home price appreciation, said UBS. If these errors had suppressed the prices seen in the recent refi wave, the latest quarterly home price appreciation number is not quite accurate, and would misleadingly show a marked spike. Further, the numbers also suggest that home price appreciation during the two-year refi wave had actually been stronger than previously indicated.
UBS explained that the 4.2% quarterly appreciation - which is not annualized - is roughly 3% stronger than recently seen levels. If this percentage were "spread out" from 2001 to 2003, it would mean that home price appreciation during this period was actually 1% greater.
"Quarterly readings are inherently noisy, but if even a fraction of this simple-minded estimation is applied, then estimated home price appreciation goes up substantially," wrote analysts. The firm's long-term models suggest that 1% more home price appreciation can boost discount speeds by 0.5 CPR
Amy Crews Cutts, deputy chief economist at Freddie, said that the higher interest rates in the last two quarters of 2003 did little to lessen interest in the housing market as consumer confidence was boosted by the economic growth in the second half of the year as well as the tax breaks that households received.
Moreover, there was spillover from June's 45-year interest rate low. The date of origination is recorded, and not the point of actual sale, which happens about 60 days prior to the date of origination. Thus, loans of borrowers who wanted to take advantage of rates in June actually might have been originated in the fourth quarter.
Crews Cutts also explained that borrowers who refinanced in December were not serial refinancers, thus they usually got new appraisals that reflected the mark-to-market value of the property. Aside from this situation, there was also an influx of purchase loans that needed a fresh appraisal, and these also reflected mark-to-market home values. These deals represented a more accurate portrayal of home prices during the period.
In contrast, many borrowers who refinanced in June were probably serial refinancers who had refinanced as early as six months beforehand. These borrowers usually underwent a streamlined refi as lenders would rather take the risk than spend another $300 on a property that was just recently appraised. These loans that did not go through a new appraisal would still reflect home values from the prior loan so if there was any price appreciation on the property, it would not show. This could have had the effect of dampening the home price appreciation numbers during the refinancing wave.
"I think the true measure is the 8.4% year-over- year growth, which is probably a much more accurate view of what has happened to house prices over the year," said Crews Cutts. She added the numbers in the Freddie Index were probably depressed due to the rapid refinancing and lack of true appraisals, which changed when loans started being marked-to-market.
However, she said that even the 8.4% growth is considerable, especially if the borrower views house ownership as a comparable investment, where a dividend is equal to one's rent plus the additional equity growth in the asset. She expects the house price Index to show more moderate growth going forward, with a projected year-over-year growth of 6% to 7%. The decrease will be due to an expected slight rise in interest rates.
Rates are not expected to rise drastically. Most believe the Fed will hold back this year, with the employment numbers projected to remain lackluster. Further, if the Fed doesn't raise rates by June, it probably won't happen this year because of the upcoming presidential elections, Crews Cutts said. "If it was any other year, the Fed may have contemplated a move in October, but if the Fed doesn't raise rates by June, then it becomes an election year problem."