American statistician Edwards Deming once said "change is not necessary as survival is not mandatory." The recent HousingWire column on the upcoming foreclosure disasters is not wrong but only if the current processes continue.

However, the robo-signing problems are not just clerical errors. They are systemic indicators of a process doomed to failure. As long as the industry doesn’t care and doesn’t change these processes to validate their reliability, any incident has the potential to result in more regulation and more lawsuits. If we continue to see change as unnecessary we will have problems surviving, as the article elegantly pointed out.

I have petitioned the industry for years for a change in process controls as well as the development of a “ranking system” to acknowledge the best and most reliable origination organization and servicer. So, while the article accurately pointed out the very real potential for further problems, it failed to offer any possible alternatives.

I think this is a disservice to readers. We have had too many doom and gloom discussions. We need some discussion that can lead to positive change.

Mortgage lenders and servicers are no different than any other company out there. They offer a product/service, which they then have to produce. This is accomplished through a series of processes. Each of these processes has an expected outcome, which produces the product offered.

There are however, some slight variations in what service industries offer and even more so when the service is a mortgage loan.

We have a bifurcated customer base. In other words we are trying to serve two masters; the applicant/ borrower and the investor.

Our product is risk and therefore our “raw” materials are the customers and their data. In the initial stage we are trying to conform the borrower’s data and corresponding risk to the investors” specifications. In the servicing stage we are working to ensure that the investors get paid and that we provide the services we promised to the borrower (i.e. tax payments, etc.). It is important to note here that while the origination process is a continuous flow approach, the servicing process is a series of activities that can occur independent of each other.

Whether continuous or independent, all processes have an output that impacts the overall performance of the company. All of these processes need to be measured to make sure they are working as expected.

If so, leave them alone, if not, fix them. To accomplish this requires several key actions:

a. The actual relationship between a process variance (mistake) and its impact on how the loan/process will perform must be measured and validated. Right now we have no idea what mistakes, if any, impact loan performance. Take fraud for example, everyone says that all lenders have some level of fraud in their portfolio and that we all accept that because it doesn’t impact performance. If that is true, then what fraud is not harmful? Do we have any idea? Of course not. Why can’t we change our processes for this?

b. That every file must be reviewed to identify what mistakes were made and a “score” provided to measure the risk of non-performance of that loan due to these mistakes. This is the performance of the process.

c. That it be done in an automated fashion so that any subjectivity is removed and the result is objective and valid.

d. This “score” can then be used to rank lenders and servicers. This ranking could be done on an overall basis or specific products, processes, etc.

Imagine that a lender wants a servicer for a large population of jumbo loans. They want a customer service process that is focused on communicating with the borrower, helping them out with issues. Because these are such high end borrowers, the risk of non-performance has shown to be minimal so when selecting a servicer they want one with a proven record of excellent customer service, not one focused on getting distressed loans to perform. Using this ranking approach they would seek out a servicer with scores that consistently are above the industry average in customer service requirements.

A program like this would take a few months to get it in place but the best part is that it can address the agency’s LQI issues and much of the associated costs; it could eliminate the costs associated with the current QC process (which by the way is a 19th century approach), it could address the “crisis” we are facing today because this process would have a score associated with it so that it would be very easy to see who is doing it correctly and who is not, and be used to reward better performers (lenders and servicers), with better pricing. It would also be “real time” results; not a point in time review which is invalid the minute the reviewers walk out the door.

Of course the down side is that your problems are exposed: sloppy process controls are easily identified: management would have to take responsibility, and there would be jobs lost. Right now the down side wins because everyone is afraid of change. But I keep quoting Dr. Deming “Change is not necessary as Survival is not mandatory."

Rebecca Walzak is president rjbWalzak Consulting, Inc.

Have an issue you want to sound off on? Email the editor of HousingWire.