M:Qube is about to launch into the mortgage market but it is no ordinary lender – it’s a fintech business first and a mortgage lender second. Joanne Atkin caught up with CEO Stuart Cheetham to find out why M:Qube is different
Stuart Cheetham and Richard Fitch
M:Qube describes itself as a fintech mortgage lender using data and technology to deliver a legally binding mortgage decision in minutes instead of weeks.
The Guildford-based firm says its mortgage platform is “more technologically advanced than any other in Europe”. It uses advanced analytics, artificial intelligence and machine learning, neural networks and proprietary technology and intends to bring mortgage lending into the 21st century.
M:Qube’s proposition is very different to other mortgage providers as it is both a mortgage lender and a mortgage technology provider. It will be ready to launch by the second quarter of 2020 starting with individual buy-to-let mortgages followed by portfolio landlords a month or so later then will expand into company SPVs.
Phase two later in the year will see M:Qube launch residential mortgages and it is eyeing up providing 30-year fixed rate mortgages after this.
Also, later in the year M:Qube will move into providing its technology to other lenders in the form of Lending as a Service (LaaS).
The beginning
Back in 2017, M:Qube was founded by chief executive officer Stuart Cheetham, formerly of Lloyds Bank and Prudential, and chief operating officer Richard Fitch, previously with Halifax and Credit Agricole. They both knew the mortgage application process was cumbersome and felt there was a better way to do this by using advanced technology.
Stuart Cheetham, co-founder and CEO of M:Qube, explained: “The mortgage process from application to offer is bureaucratic, opaque and a relatively slow process given modern standards of technology so we looked to address that.
“We wanted to be able to provide the best possible outcome for the customer. I use the analogy of Netflix. Has Netflix reinvented TV? No, but it has reinvented the way we consume TV. I am in control of what I watch, how much I watch and when I watch it. This is how we have been thinking about the mortgage process.
“Have we reinvented the mortgage process? No, because it is highly regulated but we look to give certainty and control through an adviser to the consumer so we are helping them to get their mortgage as fast, or slowly, as they want it.”
So where does the name M:Qube come from? Stuart said they wanted a name that resonates with their DNA and M:Qube does that for two reasons – it gives a sense of data and mathematics: “As a business we are a data, AI, fintech first and a mortgage business second. The cube also relates to our three core stakeholders who are consumers, mortgage advisers and buy-side investors.”
Automation
M:Qube says it has moved from digitisation to automation. As the broker fills in the mortgage application form, M:Qube sees the form being filled in at the same time so, by using AI and APIs its systems are validating the data and it can immediately highlight any discrepancies or missing information.
Tony Connell, chief product officer, with a tech background at FICO, Visa Europe, Experian and Royal Mail, explained how the system is a data driven process with real time decisioning: “First, we assesses the property the borrower wishes to buy.
“The system accesses 140 points of data on a property using various sources such as four AVM providers, the Land Registry, Rightmove and Whenfresh to gain a myriad of information. This would include the method of construction, square footage, number of rooms and floors, flood and subsidence risk and so on.
“This gives a clearer picture as to which lenders would lend on the property and if indeed it is mortgageable. Data on the borrower is then collected, categorised and verified followed by real time underwriting.”
Stuart said the company takes a very prudent risk position and uses a ‘four-eyes’ underwriting check – two machine ‘eyes’ and two human eyes – and has proven highly effective at mitigating human error.
Documentation
Customer documents supplied by the broker will be used to instantly retrieve thousands of data points relevant to an application. Because the application is happening in real time, the technology allows brokers to verify the accuracy of customer data before the final application is submitted.
Documentation, such as ID and bank statements are provided to M:Qube in a secure link. The data is tagged and independently checked via APIs linked to various third party sources.
If customers don’t want to use open banking, they can supply bank statements but Stuart says these are easy to forge in Photoshop so it can be hard to know if they are genuine or not. But the systems can spot fakes.
He said: “Our platform extracts the bank statement in a millisecond, extracts data points and checks that the bank logo, the customer’s name and address, account number and sort code are all in the right places. APIs check all this and verify if it is a live account.
“Data on the bank statement such as the mobile phone company, credit card provider, council tax, mortgage lender etc is matched against credit reference agency data. We can then be confident it is a bank statement because we have independently verified the information.
“We take five to seven times more data than a traditional lender will do and instantly assess it using deep AI technologies.”
Legally binding offers
M:QUBE says its mortgage offers are legally binding as opposed to being a decision in principle which it can do by using AVMs. If the AVM comes back indicating a physical valuation is needed M:Qube aims to deliver this in a 24-48 hour window and in this case the offer is made subject to a physical valuation.
A legally binding offer gives a buyer more control said Stuart: “As advisers supply information to us we can assess and underwrite the information in real time and that means you can have certainty and control, which has a number of benefit for consumers.
“They can start acting like a cash buyer and are able to get discounts on average of 3% to 5% off the price. They can move faster and have more control. This is our vision.
“Twenty years ago car insurance was bought from shops, now 90% is bought almost instantly online, this is the journey we want mortgage customers to go through with advisers.”
Partnerships
The other string to M:Qube’s bow is its work with buy-side investors – this could be an existing lender (bank, non-bank or building society) or it could be an insurance company or asset manager.
As well as being a lender M:Qube will offer its technology to other lenders who want to service their own loans and own the customer. That way they can use their balance sheet to provide a good price but have access to advanced technology. M:Qube said it is in conversation with existing lenders to help them innovate their journey and transform themselves.
Stuart said: “We see ourselves as a transformational partner. We can deliver better customer outcomes if we work with the ecosystem and help lenders transform. We have large and deep capability in AI, machine learning and data analytics; typically this skill set does not sit in the right form in many lenders.”
M:QUBE will be partnering with a select number of high profile financial institutions, to supply the debt capital used to issue mortgages.
Stuart added: “If an insurance company or asset manager wants to buy mortgage assets, and lots of them do, we will offer them a service where we will be the lender of record and do all the credit and servicing for them because they don’t have the capability to do that.”
30-year mortgages
With the capability to analyse such a large data set M:QUBE can provide applicants with tailored products for their specific needs. For example, the government has raised the issue of long-term fixed rate mortgages including 30-year fixed rate products. This is normal in the US but could it work over here?
Stuart believes the answer is yes and he believes they will be available in the market by end of this year or early 2021. He says the key benefit of a 30-year fix is that the interest rate and the contractual tenure of the mortgage are matched so the consumer has certainty of rate over the entire life of the loan.
Stuart commented: “In the UK more consumers are now taking out 40-year mortgage terms and a two or five-year fixed rate so they are mismatching their contractual tenure. If interest rates go up they are exposed to a risk of higher payments in the future.”
Another advantage is that the 3% stress test that lenders have to use when assessing customer affordability would not be applicable so borrowers would be able to take out a higher mortgage.
Stuart explained: “If you match the contract tenure and the fixed rate tenure together there is no need for a stress test under the FCA responsible lending guidelines so affordability goes up. We have calculated that you can typically offer a loan of maybe 30% more but still be very clear that it meets affordability guidelines.
“That product is not sold in this country because the big banks want to sell five-year rates. But insurance companies want 30-year mortgages because they have pension annuity funding.
“So if we can build new products with new funders and offer the ability to distribute that into the UK, consumers get a completely new type of product which offers, for some people, a different and potentially better outcome.
“We are talking to a number of insurance companies about doing exactly that and there is huge interest from these companies.”
