Brokers complain some more

Wednesday, March 3
Written 10:30 a.m. EST

RATES COME TODAY: Today is the day when Bankrate's research department conducts the weekly mortgage rate survey. Last week the 30-year fixed averaged 5.15 percent and I think it'll fall to 5.11 percent this week. For most borrowers, this won't reflect much of a change in rate and fees.

Keep in mind that the February employment report comes out Friday morning. As I explained yesterday, investors are ready to ignore any news in the employment report that normally would result in lower mortgage rates. So if the employment report influences mortgage rates, it will be to force them upward. That means it's risky to float beyond Thursday afternoon.

The consensus is that the report will say that nonfarm payrolls fell by 20,000 in February. Normally, a bigger-than-expected drop in nonfarm payrolls would result in lower mortgage rates. But if nonfarm payrolls fell a lot more than 20,000, then investors will chalk it up to snowy weather. They'll look upon the job losses as a temporary blip. As a result, mortgage rates are unlikely to fall on the poor economic news, as they normally would.

The Bureau of Labor Statistics will try to account for the snowstorms in the northeast, in an attempt to arrive at a relatively undistorted nonfarm payrolls number. I doubt that investors will realize this, or trust in the BLS's accuracy.

LOTTA VENTING: Monday, I let mortgage banker Alan Rosenbaum vent here about the state of the mortgage business. He's fed up about paperwork requirements that he believes impose more costs than benefits on consumers and lenders. He agrees with the goal of consumer protection, but thinks the feds fumbled on the details. And he contends (and I agree) that federal policy unfairly promotes the biggest lenders to the detriment of smaller lenders and mortgage brokers.

Mortgage broker (and RTI participant) Michael Becker agrees in a long, passionate e-mail to me. He was glad to hear that the federal housing department would revamp the Good Faith Estimate form "to try and prevent the bait-and-switch."

"Unfortunately the rules they came up with treat brokers and lenders differently, complicate the process, and I believe may encourage fraud," he writes.

He adds: "The Good Faith Estimate is now three pages long and many originators still aren't sure how to prepare it correctly. On purchases, seller-paid closing costs are supposed to be disclosed on the Good Faith Estimate. This has the effect of giving you an incorrect dollar amount on the cash to close requirement, complicating underwriting. Your article mentioned a 10 percent variance in third-party fees, but did you know that you have no variance for some fees like origination and recordation or transfer fees? If you are wrong about government fees it will come out of your compensation."

That's especially a problem along the East Coast, where states, counties, cities, towns, townships and even special agricultural districts levy taxes and fees and stamps. In New York, there's a mortgage recording tax that varies, depending on location and even the number of units in a building. It's easy to make mistakes, and brokers and lenders are responsible for the costs. I'm not sure that's fair.

Take, for example, someone who borrows $1 million to buy a co-op in New York City. If the building has three units or fewer, the mortgage recording tax is $18,000. If the building has four units or more, the mortgage recording tax is $25,500. This complicated system invites mistakes. I can easily see a situation where an error leads to a GFE that says the mortgage recording tax on a million-dollar loan is $18,000 when it should be $25,500. The broker or lender would be responsible for paying the difference. I don't think that's fair.

Becker writes: "It would have been easy to stop the abuses that RESPA reform was trying to stop without such a complicated system. First, to stop excessive fees on loans, why not cap compensation on mortgages like they are on reverse mortgages, where you are limited to a 2 percent origination or $6,000, whichever is higher? To stop the bait-and-switch tactics, all you would have needed to do was to make a very simple rule that rates can't be raised without being disclosed two weeks before closing and without a valid reason, like the LTV being higher or the credit score dropping, and requiring closing costs (outside of prepaids for interest, taxes, and insurance) to come within 10 percent of the amounts listed on the GFE or the difference comes out of your compensation."

He wrote a lot more, and I might share some of that with you later.

WRONG OR NOT?: Yesterday, I said that one flaw of the Home Affordable Refinance Program is that it doesn't allow borrowers to be removed from the note -- for example, in a divorce. I said HARP doesn't allow borrowers to be added to a mortgage, either.

A reader writes: "The correct information is as follows. Fannie Mae and Freddie Mac do allow people to be removed in case of divorce or death. For divorce, a divorce decree is required. In the case of death, a Death certificate is required. You were partly correct about adding borrowers. Freddie Mac does not allow borrowers to be added. Fannie Mae allows borrowers to be added for any reason."

After spending a ridiculous amount of time digging, I found a Fannie Mae document (warning: PDF) that says that Fannie's Desktop Underwriter software "requires the borrower(s) on the existing Fannie Mae loan to be included on the DU loan application." It doesn't say anything about exceptions for divorce decrees. And indeed, it says a borrower can be added to a Fannie-owned loan in a HARP refi.

I've run out of time to find the Freddie guidelines. If anyone wants to send them my way, have at it.

Read more mortgage blogs.


Show Bankrate's community sharing policy
          Connect with us

Timely market news and advice for consumers ready to buy, sell or invest in real estate. Delivered weekly.


Crissinda Ponder

Homeownership falls to level last seen during Beatlemania

The national homeownership rate is at its lowest level in 48 years, according to new data from the U.S. Census Bureau.  ... Read more

Partner Center

Connect with us