My New Blog

In a recent article, I examined Bank of America's new no-fee program for house purchasers, under which lender and third-party fees are absorbed by the bank. On a fixed-rate mortgage, the borrower pays the interest rate and points, and that's it. Price shopping would be so much easier, I mused, if all lenders did the same.

A spokesman for another large lender responded to that article by claiming that other lenders do offer the same product, but they give it a different name; they call it a "no-cost mortgage."

I will explain what he means with an oversimplified example. A lender who absorbs all costs in its rate and points must mark up the price accordingly. For example, if BofA is prepared to absorb $3,000 of costs including third-party charges to acquire a $300,000 loan at 6 percent, it will price a no-fee loan at 6 percent and zero points.

Now consider Lender X offering the same loan, and faced with the same $3,000 in costs except that the costs are billed to the borrower. This lender offers 6 percent at -1 point, which is a rebate to the borrower. The borrower selecting the 6 percent mortgage on which the rebate just covers the $3,000 has a no-cost mortgage from X that appears identical to that of BofA's.

In other words, the borrower from BofA pays 6 percent and nothing else. The borrower from X pays 6 percent plus $3,000 in fees but receives a $3,000 rebate from X.

But note a critical difference. The $3,000 charged the borrower by X is probably not guaranteed. A few lenders guarantee their own fees (see below); even fewer guarantee third-party fees. All the rest provide estimates, which sometimes have a funny way of escalating as loans move toward closing. So the borrower dealing with X might end up with a rebate worth $3,000 and be billed for $4,000 in costs. That can't happen with the BofA mortgage.

Most borrowers are not aware of the no-cost option from lenders other than BofA. The loan officers and mortgage brokers with whom they deal are unlikely to volunteer the information because no-cost loans are easier to comparison shop. If the borrower requests a no-cost quote, they will comply, but the quotes are based on cost estimates that can be far off the mark.

Borrowers can roll their own no-cost mortgage online. They do this by selecting an interest rate that carries a rebate large enough to cover the settlement costs. This requires that they have access to the complete range of rate/point combinations offered by the lender, as well as the settlement costs.

Unfortunately, very few lenders provide this information on their Web sites. Among those that do are Upfront Mortgage Lenders (UMLs), which are listed on my site. UMLs must provide this information in order to be certified.

As a source of no-cost loans, UMLs have advantages and disadvantages relative to BofA. UMLs provide online pricing for a wider range of products, and they provide complete pricing data on adjustable-rate mortgages online, which BofA does not. On the other hand, BofA pays third-party fees as well as its own, whereas UMLs guarantee only their own fees. Third-party fees are estimates -- honest estimates, but still estimates.

NOTE: If you expect to have the mortgage five years or longer, you don't want a no-cost mortgage unless you are desperately short of cash. If you have the cash, it pays to buy down the interest rate by paying points rather than the reverse. This calls for a revision of your shopping strategy, from finding the lowest no-cost rate to finding the lowest cost at a specified rate below the no-cost rate.

For example, if a 30-year fixed-rate mortgage with no cost is available at 6 percent, set 5.5 percent as your shopping rate and find the lowest cost at that rate. With BofA, it will be the points charged on the 5.5 percent loan, which may or may not be available online. With UMLs, it will be the sum of lender and third-party charges at 5.5 percent, which will be available online.

The writer is professor of finance emeritus at the Wharton School of the University of Pennsylvania. Comments and questions can be left at www.mtgprofessor.com.

August 2007

Solds decreased by 14.63%.

As for Average Prices:

The "New Listings" average list price is up 14.56% to 299,055.

Sold average sales prices increased 8.33% to $256,035 compared to $236,355 in 2006.

Numbers compiled by Alamo Title, Austin

Damned lies and median house prices

Inman News September 24th, 2007

One of the most commonly referenced statistics in the real estate industry is the median sales price of homes. Many articles are published in newspapers the day after the National Association of Realtors releases its quarterly Metropolitan Area Existing-Home Prices report with conclusions about the healthy or unhealthy state of the local real estate market. Unfortunately the median sale price is frequently taken out of context and misinterpreted.

Even someone who slept through most of their math classes will remember that the arithmetic mean and median statistics are different for the same data set. The average is the sum of the numbers divided by however many numbers you started with. The median is the number in the middle, when the numbers are listed in order.

The reason that the mean or average sale price for a market area can be misleading is intuitively obvious and that’s why it’s rarely cited. A few sales in the extreme luxury segment of a market — think $30 million or more in Bel Air — can push up the average for the entire local market area.

The median sale price can also be misleading particularly in down markets. Let’s consider the San Francisco market. The median home price reported by NAR in the first quarter of 2007 was $748,100. In the second quarter it was $846,800 — a jump of more than 13 percent. Wow, that appears to be a sure sign of a healthy market. Or is it?

To get a truer picture of market conditions, let’s consider a few more statistics: price indexes (using a repeat-sales price methodology), the number of sale transactions, price reductions and inventory growth. From April to June 2007, the S&P Case Shiller Home Price Index showed a decline of just less than 1 percent, not an increase of 13 percent. Likewise the home-price index published by OFHEO showed a decline of just less than 1 percent for the second quarter. The S&P Case Shiller index covers only resale transactions while the OFHEO data covers multiple sale types but only for conforming loans.
…
So what actually happened? Sales transactions increased with a greater proportion on the high end versus the previous quarter. There appears to have been little actual appreciation as evidenced by the Case Shiller and OFHEO numbers, while inventory increased and prices of many listed properties were reduced. So next time you read that median house prices have increased in your area, don’t celebrate prematurely. Conduct more research before you reach a conclusion about market conditions in your area.