The overhead of delivering better consumer experiences, shorter and lower cost turnaround times whilst dealing with the raft of increasingly stringent regulatory requirements is making the mortgage lending space feel more frenetic than ever before.

If you couple that with low rates and new FinTechs entering the market leaping legacy technology challenges, the task at hand can seem almost overwhelming. 

CoreLogic is fortunate to be at the heart of the interactions between home owners, brokers, lenders and real estate agents and has made it our mission to help Australians and Kiwis find, acquire and protect their homes through the power of property data, connectivity and analytics.

As such, we have been working to help empower our customers to solve for some of these challenges and, where possible, alleviate some of the regulatory burden.

The three regulatory standards we are commonly asked to support are APG223 (Residential Mortgage Lending), APS220 (Credit Risk) and APS112 (Capital Adequacy).

These are each lengthy documents with many requirements.  Summarised below are several of the most common ways in which we are helping our clients solve for some of these challenges with systematic approaches using data and analytical insights:

1. Appropriate valuation strategies, lender and broker adherence to policy and motoring of third party providers models 

All 3 APRA regulations address (to different degrees) regulatory requirements relating to the monitoring and management of:  credit risk, valuation of securitised collateral (including automated valuation models), third party brokers and introducers, and appropriate assignment of risk weighted assets.

Valuation strategies often become complex over time, making adherence to policy and maintaining transparency of process increasingly challenging. 

Consistently, we see lenders having difficulty in systematically managing geographic risks and lending concentration, particularly at an individual apartment development level. We also see various different capabilities of reporting on how lenders are adhering to documented policies. The AI & Machine learning steamroller has also caught Automated Valuation Models, which is driving great innovation from an accuracy and coverage perspective, but brings nuance into fitting the guidelines of APG223 in appropriate use and stringent monitoring controls.

Here are some ideas for areas that can help:

  • Independent reviews of strategy and analytical simulation of valuation outcomes can give excellent insight for instant improvements to meet the regulation and find areas of potential optimisation to deliver faster turnaround times.
  • Consider a transparent data driven approach to set postcodes of interest – this helps with both transparency of method and also an automated refresh of geographies 
  • Set geographic and development level lending thresholds that can automatically trigger management intervention 
  • User level behavioural reporting on manual overrides to policy – also managing valuations that do not convert into loan applications
  • Ensure providers of analytical models have the appropriate governance and monitoring frameworks in place

2. Ongoing monitoring of the portfolio and stress testing

We see a number of different ways that our customers are conducting these exercises, ranging from portfolio level indexing to highly sophisticated security level analytics. Many customers are too constrained by costs, technology and resources to adopt the more sophisticated methods, but we increasingly see senior stakeholders and regulators requiring fast granular insights into the portfolio.

Some of the observations we see that can help with this:

  • There are multiple interested parties in the Home Loan portfolio
  • Consider enterprise alignment on best practice portfolio valuation 
  • Course indexing may be appropriate for portfolio level views but not for customer level treatment strategies or capital allocation
  • Visualisation tools are increasingly adopted to answer questions quickly
    • geographic LVR distributions
    • exposure to hazards and economic risks
    • exposure to development of interest
  • There are cost effective methods to digitally solve for many of these challenges

Beyond the two themes summarised above, on our journey to the vision of presenting our customers with a digital property DNA string to empower highly automated security level decisions, we are evaluating our customers most contemporary property level challenges. On a daily basis we are asked about exposures to developments clad with Aluminium, Mascot tower, ongoing costs of owning property and how will drones and VR will be capturing data to perform more efficient valuations.

We have some exiting developments in data liberation in the coming months, and are continuously evaluating business challenges to determine our areas of focus. If you anything challenging interesting or plain out there related to property we would love to hear from you.