Loanlogics finds a persistent error percentage of 11.5% in American mortgage files
Based on the analysis of industrial data and after evaluating thousands of lenders, Lowlogics estimates that inefficient systems, errors of loans and resulting delays during the mortgage process have been translated into around $ 7.8 billion in higher costs for consumers.
Loanlogics investigated data for 2014, 2019 and 2024 and evaluates “DOC to Data” discrepancies where information is or is incorrect in a file. The “Doc to Doc” discrepancies also assessed the documentation that claimed part of a file is or is incorrect.
“Our results show zero material improvement in the quality of the loan file after a decade of industrial investments and assumed innovation,” said Craig Riddell, Executive Vice President of Market Development at LanLogics.
“A common challenge with recent technology investments is an inappropriate application. Poor results and redundancy are the by-product of incomplete or poor training, hurried implementation and aging integrations, which leads to unexpected and expensive data conflicts that require manual intervention,” he added.
It is remarkable that the combined error percentage for DOC-Data and DOC-to-DOC transfers increased from 9.7% in 2014 to 13.3% in 2019, before he fell to 11.4% in 2024.
“The peak in error percentages in 2019 correlates with higher mortgage volumes in the entire industry. This was probably due to fluctuations in inexperienced personnel that was caused to tackle the increased workload,” said Roby Robertson, Executive Vice President of Origination technology strategy at Lanlogics. “Lenders responded to the reduced volume in 2024 with reductions of the workforce, which led to more experienced and expert staff, and we saw error rates fall as a result, but still close to the average of 10 years.
“As we continue to see that new approaches of creative lending come to the fore to serve more and more of the lender population, it is absolutely necessary that companies work to solve their data problems with better automation and technology,” he added. “We help everyone, from lenders on consumers to aftermarket loin buyers and securitizers, understand what is broken in their files and not only to identify problems, but also to correct.”
Why is the error percentage still that high?
Despite many investments in technology and automation in the past decade, a high error percentage indicates a need for a different remedy.
“The investments that have arisen in the last decade in the origin Housing.
“The analogy that I give is that a mortgage is still like a large cardboard box full of files and it is just a kind of moving the assembly line. And that next person opens that box and makes their assessment of what is going on, and then they say when they say they have that box, and then they are in that box.
On a fundamental level, Robertson said, whether a file goes down the line and is assessed, is up to the individual.
“That is why the costs for the production of a loan have been raised,” he said, adding that the costs are more than $ 11,000, per Freddie Mac facts.
Fouleous loans can also cause problems when it comes to buying loans. Riddell said that Loanlogics can see when a loan will contribute to the rising error percentage.
“If a loan starts to wiggle when it is in the payment problem, then that file will get a deep dive and, in combination with performance and possibly some dates errors, that is where some return activity starts to start,” said Riddell.
“There is a difference between data consistency and manipulation and fraud. They must therefore comb which types of errors are found.”
At the end of 2024, Robertson said, began to drive money lenders outside the boundaries of traditional mortgage loans.
“At the end of 2024 they pushed the edge of a traditional box and created non-QM,” he said. “It was not a deliberate non-QM strategy, but deals that fell outside of traditional lending, go Jumbo or deal with poor credit or high DTI to get the deal over the line. The industry was still shaking and tried to make every deal happen.”
In other words, Robertson said that lenders just did what was needed to take out loans. This often meant the approval of borrowers with higher than usual debt-to-income ratios, lower credit scores or loan sizes that exceed the conforming limits. Ultimately, this resulted in more standard values in 2025.
But Robertson believes they will relieve.
‘Lenders noticed it [the defaults] Immediately. I think the data they get from these portfolios are just much better than before. So they recognize immediately, and they actually shift to more smart non-QM deals. They go behind non-QM deals that are logical. “
