Are We On the Verge of a Lending Surge?
Coming off an election cycle, pandemic in the rear-view, inflation finally easing, rates going down for the first time in years — there’s a moment of financial release that’s waiting to happen. Banks and lenders need to be ready for this convergence of opportunity with technology that supports rapid data processing and decision-making.
By Anand Pandya, Global Head of Financial Services at Hakkoda
It’s not breaking news to say that financial markets have been transfixed with the global economy’s biggest lever getting pulled one way or the other this November. Everyone knows that significant decisions have been delayed or suspended in a prolonged holding pattern across the sector — with the notable exception of mortgage lenders.
Back in mid-September, the Fed cut its benchmark interest rate by 0.50 percentage points, marking its first rate reduction in four long years. This knocked the federal funds rate into the 4.75% to 5% range, down from 5.25% to 5.5% (which was its highest level in over two decades). Retail bankers and mortgage lenders have been abuzz ever since with the rallying cry of "Refi now!"
The Fed went hard at 50. Institutional interests are still generally in "wait and see" mode and very much hoping the rate will go even lower, as another 25 bps cut came November 7 followed by another on December 18. There may be more in 2025. In coming months, this could really trigger pent-up demand and spark a dramatic surge in lending activity.
With no disrespect to death and taxes, you can be certain that when rates go up, volume goes down. And when rates go down, volume goes up.
Check Under the Hood
Lenders and banks are doubtless doing the math and anticipating engagement since their entire world is about the cost of lending. What they’re probably not thinking is "Gee, I ought to migrate my databases."
They should be.
From a technology perspective, when anticipating an onslaught of new volume, you should know whether your systems are going to be able to handle it. According to the MIT Computer Science and Artificial Intelligence Laboratory, "more data is now produced every month than existed in total just over a decade ago." If you simply extrapolate the growth rate in data generation and consumption YoY, even systems that were perfectly up to snuff just five years ago may now falter.
A lot has changed in an awfully short period of time, and financial data architecture needs to be able to accommodate it.
AI and the Cost of Lending
Remember when the relatively unknown nonprofit OpenAI released a "free research preview" of a large-language model update and a little tool called ChatGPT? That was only two years ago.
Now consider the associated data science advancements and AI tooling already incorporated into every facet of the banking business — whether that’s pricing, production, risk, underwriting, servicing, or compliance—their thirst for data has grown exponentially.
Is the average legacy SQL system capable of supporting all that to deliver what the prospective customer needs from a rate drop, or is it actually going to pull down margins under the weight? It can’t have a positive impact on the cost of lending if the offset’s not there.
Read more:
- AI-Assisted Lending Could Boost Small Banks. But Regulatory Fear Stifles Innovation
- Balancing Risk and Speed in the New Era of Digital Lending
- Future of Mortgage Lending Hangs on a Balance Between Innovation and Safety
Digital Reality Check
So what does deliver the offset? Consider some situational awareness questions applied to your institution’s technological capability:
Do you have all of your data in one spot or is it still siloed across front, middle, and back-office domains? Can you conceivably know that Customer X is the same person who entered a rate calculator study on your website, that also has a credit card with your bank, as well as a checking account, and has previously paid off an auto loan; or does that Customer X profile live in five different systems with five different ideas about who that customer is?
There are things you should understand "digitally" about your customer and her behavior patterns, not in a creepy Big Brother way, but simply because you’re her bank in the 21st century. She has kids and is the fiduciary on their savings accounts with your bank. Her kids are going to be heading off to college. Hey, she’s also got a 10-year-old vehicle and she’s on the West Coast. Maybe she’s also interested in an EV?
When she calls and says, "I’m interested in refinancing," are you prepared with the right information for all these opportunities? Can you predictably say: "we should be throwing her a 50 point discount on her rate because we understand what she wants to do and we can package a simultaneous refinance with a cash out and a loan for an auto, and now we’re creating demand?" She’s already your customer. Does it need to take 60 days to do the underwriting process for her? Can we expand our Buy Box for her if her FICO’s a little short with just a tick to the rate because we actually know she’s good for it?
If you can’t do those things as quickly as you pick up the phone or service them on her timeline, guess what? You are basically telling Customer X to take their business across the street. That’s the market now.
Meeting the Moment
Coming off an election cycle, pandemic in the rear-view, inflation finally easing, rates going down for the first time in years — there’s a moment of financial release that’s waiting to happen. There are people who want to start doing things again. There are companies that want to do things again. And there are new tools that make it possible to do those things so much faster.
Banks and lenders need to be ready for this convergence of opportunity with technology that supports rapid data processing and decision-making.
So yes, you should be thinking about your databases — and your entire data architecture — with a clear line to strategic enablement, where every dollar invested is expected to drive discernible business value and enhance customer experience. If you don’t meet this rare moment, somebody else will.
