CAPE Analytics has unveiled a new, AI-powered, automated property condition report (aPCR) tool crafted for institutional lenders, whole loan investors, and real estate investors.
This latest product uses computer vision to assess aerial imagery from airplanes and other sources and extracts information about property conditions, with a focus on marketability and property value, the company said Wednesday.
CAPE relies on multiple imagery capture partners, which allows it to pull together novel datasets. The solution renders an array of nationwide property information available in one PDF report. It makes accessible information that used to be hard to obtain, such as the condition of the backyard, the proximity of a property to a loud road, train tracks, or a body of water as well as the occupancy status of a property. The data can also be accessed via an API, in bulk data, or through a dynamic web application.
It allows home equity lenders, whole loan investors, and single-family rental investors to get “objective” and “current” exterior property condition assessments, with a focus on changes that can affect property value, the company said.
The CAPE aPCR tool can be applied across the valuation spectrum—from adding condition validation to an automated valuation model (AVM), to driving appraisal workflows. The efficiency of this solution also comes at “a fraction of the cost of traditional methods,” said Jiapei Wang, senior product manager at CAPE Analytics.
The mortgage ecosystem evolves quickly, with different sub-sectors such as lending, loan aggregation as well as loan securitization being impacted, Sean Begley, senior director of business development at CAPE Analytics, told HousingWire. The common understanding of valuation, inspection and even property value is shifting, as more and more technology-enabled alternatives are emerging. Begley said the CAPE Analytics aPCR represents that missing piece, instantly informing mortgage professionals about conditions. The tool can be paired up with other types of valuation metrics.
“The real competition for us is the status quo,” added Begley.
However, some elements of images, such as angles, can be deceiving, Begley acknowledged. When it comes to external obsolescence factors, such as the proximity of a high tension power line, it can be hard to evaluate the exact distance between the property and the potential danger. To address the issue, CAPE Analytics developed what it says is a reliable method, calculating the distance from the nearest point on the parcel to the object of interest to return an objective, unbiased measure.
“Our approach here is an example where we bring the best datasets to solve client problems, so we use multiple datasets to build out this capability,” said Begley.
CAPE Analytics said existing human-driven, visual inspections like traditional property condition reports (PCRs) miss 70% of property issues identified by the CAPE aPCR.
The product, which is now available, comes following major pushes from federal agencies to adopt automated valuation models. Fannie Mae in late April said it updated its selling guide to note that appraisals were no longer the default valuation standard.
Two months later, six federal agencies requested comments from the public on a rule designed to ensure the credibility and integrity of models used in real estate valuations. The proposed rule would implement quality control standards that govern AVMs used by originators and secondary market issuers in valuing the real estate collateral securing mortgage loans.
The mortgage industry, however, has expressed some reservations about “unintended consequences” of new quality control standards for AVMs.
In a letter to regulators, the Mortgage Bankers Association and Consumer Bankers Association said that AVMs and technologies like them can alleviate appraiser shortages, reduce transaction costs, and safeguard against individual appraisal bias. Ultimately, a robust regulatory framework continues to be a critical imperative to achieve these outcomes.
Any regulation, they argued, should consider the practicalities of model risk management and its potential unintended consequences.
