The End of Mortgage Brokers, Part 2

I got some very interesting responses to my blog entry yesterday on mortgage brokers, in which I basically said that they were parasites on their way out who wouldn't be missed.

Ken Houghton reckons that there must be some good mortgage brokers out there - and that Countrywide, for one, has enough data to be able to tell the good from the bad. Which is probably true - and I doubt that Bank of America is going to dismantle the entire mortgage channel overnight. They're more likely to simply let it wither and die: cut off the bad brokers, allow the good brokers to keep on going, but not accept any new ones.

Then again, Ken doesn't seem to have talked to Thomas, who found that every broker he talked to wanted to put him into a property he couldn't afford. Which would align with the brokers' incentives: the larger the mortgage, the larger the commission.

And PaulD made the point that what we're seeing with mortgage brokers uncannily mirrors what we've already seen with the disintermediation and demise of travel agents. PaulD misses his travel agent; I don't. I guess it's a matter of taste, as well as a function of how good your travel agent was.

And then a mortgage broker, Susan Duffy, wrote to me saying she simply didn't believe that JP Morgan's default rate on brokered mortgages was three times higher than its default rate on mortgages originated in-house. How could it be, she said, when the same team was doing the underwriting for both channels?

I looked into it, and found this presentation from JP Morgan, dating from November 2007: it's reasonably up-to-date. Check out these two graphs, showing what happened to broker-originated loans between the end of 2006 and the end of September 2007:

The difference, it's fair to say, is pretty startling. But then again, it turns out that comparing broker-originated loans to loans originated through the Chase retail channel is not really comparing apples to apples. While the credit scores of the borrowers in each channel were pretty much identical, the loan characteristics most decidedly were not. The Chase bankers' loans averaged 68% of the value of the house, and only 4% of them breached the 90% level; the brokers' loans averaged 82% of the value of the house, and fully 31% of them had a loan-to-value ratio of more than 90%. What's more, 44% of all broker-originated loans came from either California or Florida, compared to less than 1% of the in-house loans.

So maybe it wasn't completely the brokers' fault after all: maybe it just so happened that they were in California and Florida, financing speculative purchases with little or no money down, while the Chase bankers were dealing with much more conservative customers.

All the same, mortgage brokerage is a fast-dying industry, whether you like it or not. It makes sense: one of the reasons there were so many brokers in Florida and California is precisely that those states were where the brokers were in highest demand. Now that the demand for their services has fallen off a cliff, both from borrowers and lenders, the industry will naturally wither away.