Magazine article Mortgage Banking

A Buyback-Prevention Plan

Magazine article Mortgage Banking

A Buyback-Prevention Plan

Article excerpt

All lenders today share the same concern: being asked to buy back a loan--one that an investor has rejected because it was erroneously approved. [paragraph] The aftermath of the mortgage lending industry meltdown and the subsequent recession significantly expanded the potential for buybacks and other borrower credit-related problems, due in large part to errors made during the underwriting process. [paragraph] My company recently observed this problem first-hand when a lender client asked for guidance regarding a loan that he felt was in danger of falling into the dreaded buyback category. [paragraph] It turned out that by being rushed and not comparing a borrower's credit report and the mortgage application form 1003, an underwriter missed a critical red flag--that a customer applying for a new purchase loan already had an existing mortgage that was undergoing an interest-rate modification. The underwriter approved the borrower, the loan closed, and subsequently the investor advised the lender to be prepared for a repurchase. [paragraph] While we helped this lender address a troublesome individual situation, that specific transaction was another warning sign that the mortgage lending industry faces a significant but solvable challenge. That challenge can be summed up this way: Dealing with a massive amount of data has made consumer credit analysis much more complicated.

The shift

It wasn't that long ago that the process for evaluating a borrower's creditworthiness was much simpler and less conducive to errors. An underwriter carefully analyzed the credit report, the applicant's 1003 and other available information, and watched for any red flags.

Agency guidelines and investor rules were relatively straightforward, albeit numerous. It seemed to be a clear-cut process that still allowed for the underwriter's human touch.

Of course, automated systems designed to streamline and make the process more uniform didn't eliminate the need for the personal touch.

In 2007-2009, as the standard procedures and protocols became more relaxed, credit blemishes were overlooked and borrowers who previously were borderline (at best) had their loans approved.

Then, as a result of the lending industry crisis and recession, a combination of events occurred that made credit analysis and approval more complicated for lenders and their underwriters.

A key factor was the proliferation of new rules issued by the Federal Housing Administration (FHA), Fannie Mae and Freddie Mac. In addition, the number and diversity of investor conditions multiplied and there was a lack of consistency among them.

During the last few years, short sales and unique loan modifications became more common, as have borrowers with a previous foreclosure or bankruptcy--all of which required additional guidelines for processors and underwriters to master.

Of course, credit bureaus receive their data from third parties. They rely on credit-card companies and other sources to provide accurate information, and with the massive amount of new data, the potential for errors became greater.

Within a few years, even the average consumer's credit file became more difficult to interpret.

In many cases, lenders' loan origination systems (LOS) and internal operating procedures that were well suited for the pre-crisis environment haven't kept pace with the myriad changes. That includes the training programs that many lenders reduced during their recessionary budget cutbacks.

By 2010, we had the ingredients for a perfect storm--a greater inability to ensure data integrity with loan transactions.

The primary challenges

From my company's extensive discussions with lenders, title companies and others, we're seeing more confusion in the analysis and reporting of credit information. This of course fuels the potential for ignoring or misinterpreting critical data that can easily lead to unnecessary problems. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.