The Mortgage Bankers Association (MBA) expects mortgage originations to trend down to $2.57 trillion in 2004, from an all-time high of $3.8 trillion in 2003. This is part of the association's three-year forecast for the U.S. economy and housing finance market released last week.
At a press conference held last Wednesday, MBA Chief Economist Douglas Duncan said that despite these declines, "recall that total originations as recently as 1996 (including refis and purchase) were less than
$1 trillion so these are still strong numbers." He added that further drops in mortgage originations are expected in 2005 and 2006 to $1.96 trillion and $1.84 trillion, respectively.
The main source of the dip in mortgage originations is refinancing, which is dropping to an expected $1.18 trillion in 2004 from $2.53 trillion in 2003. On the flipside, purchase originations are expected to increase going forward. Purchases are predicted to reach $1.38 trillion in 2004, topping the record set in 2003 of $1.28 trillion. Increases are forecast for 2005 ($1.37 trillion) and 2006 ($1.42 trillion) as well.
In terms of mortgage rates, the MBA has mortgage rates increasing in concert with the 10-year Treasury yield, which is expected to rise to an average of 4.3% by 4Q04, 5.0% by the last quarter of next year and 5.4% by the end of 2006. Duncan said that the 30-year fixed-rate mortgage is expected to average 5.6% in 2004, but will end the year higher at 5.8%.
"With inflation under control in the presence of a growing economy, the 30-year mortgage rate will rise to slightly more than 7% by the end of the three-year forecast period," said Duncan in press release. "From a historical perspective, these are very modest interest rate increases for the level of economic growth we are expecting, and should cause few adjustment problems for the housing industry."
In contrast to 2003, when mortgage rates followed a V pattern - reaching their lowest point in mid-June and rising again at the end of the year - mortgage rates in 2004 will follow a more even pattern, Duncan said. However, the expected average mortgage rate in 2004 - 5.6% - is lower compared to 2003 when it was at 5.8%.
Existing and new home sales
The MBA also said that existing home sales would probably dip slightly this year by 1.7% to 5.995 million units from the record-setting 6.101 million units in 2003. These are expected to dip further by 6.8% in 2005. New home sales, on the other hand, will rise by 0.7% (to 1.095 million units in 2004 from 1.087 million units in 2003) before decreasing by 10% in 2005 and even further by 1% in 2006, projections indicate.
Duncan said new home sales are usually driven by move-up buyers. These borrowers usually have bigger loans, as the average price on a new home is higher than an existing home. "So to the extent that house prices have risen, borrowers in higher income categories are going to be buying up the property with the higher value, or new homes rather than existing homes," said Duncan.
He added that house prices have strongly accelerated, and affordability issues will probably impinge to some degree on first-time homebuyers who are now entering the existing home sales market. Duncan also noted that part of the purchase activity is the buying of second homes by baby boomers.
However, there are some caveats to the forecast. For instance, the magnitude by which borrowers in a move-up phase drive up home sales remains uncertain. These buyers saw an opportunity when interest rates fell this year. However, this opportunity disappears when rates change directions. Aside from this situation, many first-time homebuyers may have bought houses this year because the rate environment has been so good. These factors might make the MBA's forecast seem overly optimistic.
"On the national level, we are out of the price bubble camp," said Duncan. The MBA expects average home prices across the country to flatten, suggesting that there are some areas where the prices are dipping. He said the coastal markets are probably going to experience the greatest declines in home prices because the pace of appreciation that occurred in these regions has been much stronger. Also, these regions have had more price volatility historically.
"The problem with the coastal markets is that they have supply constraints, which means supply is completely inelastic, " said Duncan. "So what happens is small changes in demand - either increases or decreases - can lead to large changes in price." He gave the example of San Francisco, where prices fluctuated significantly after the decline of the tech boom. There are also more growth restrictions in these areas, which reduce the supply of land available for expansion.