Thank you for standing by, and welcome to the Harmoney Corp Limited FY 2024 results webinar. All participants are in listen-only mode. There will be a presentation, followed by a question and answer session. If you wish to ask a question via the phone, you will need to press the star key followed by the number one on your telephone keypad. If you wish to ask a question via webcast, please enter it into the Ask a Question box and click Submit. I would now like to hand the conference over to Mr. David Stevens, CEO and Managing Director. Please go ahead.
Hello, and welcome to Harmoney's full year twenty-four results presentation. I'm David Stevens, the CEO and Managing Director of Harmoney. With me today is Simon Ward, our Chief Financial Officer. I've been looking forward to presenting today's results for the last eighteen months, as we have some very exciting news to share with respect to our number one strategic priority. That being our new platform, Stellare 2.0, which became fully operational in Australia in July 2024. Now turning to slide two. Today, I'll begin with reminding you of Harmoney's key differentiators. I'll then take you through our key highlights for financial year twenty-four and the exciting results of Stellare 2.0. I'll then hand over to Simon, who will take you through our financial results. Finally, I'll discuss our outlook before responding to your questions.
You can submit a question on the webcast at any time during the presentation by typing it into the Ask a Question box and then hitting Submit. We'll then endeavor to respond to your questions during the Q&A at the end. Now turning to slide three, and then on to slide four. What sets Harmoney apart from others? We're Australia and New Zealand's largest 100% online consumer direct lender. The total market opportunity of 150 billion, with our market share less than 1%, so we have a huge total addressable market in front of us. Our algorithms partner in real time with Google's algorithms to attract prime, high-intent customers at low cost. We provide a great customer experience, and direct customer relationship attracts returning customers at near zero cost.
We use deep first-party data and AI models to deliver a prime loan book and a 5% risk-adjusted income. We are funded by three of the Big Four Australian banks, plus securitization programs. Our Stellare automation drives a cost-to-income ratio of 24%, and the last half represents our fifth consecutive half of positive cash NPAT and loan book growth. Just a quick reminder of our products on the right-hand side of the page. Our loans are up to NZD 70,000, with an average new loan size of NZD 20,000, which is disbursed to customers within minutes. We offer personalized rates based on borrower's risk profile. We don't charge any fees other than a one-off establishment fee, and all our loans are fully compliant with NCCP and CCCFA in Australia and New Zealand, respectively.
Our loans are typically used for renovations, debt consolidation, and helping people with life events such as travel, education, and weddings. Now turning to slide five. For those of you who prefer a visual representation of our business and the core unique features that set us apart and will drive our success, we set out a simple diagram of Harmoney on a page. Starting at the bottom left, we've spoken about the market size and total addressable market of over NZD 150 billion. We now have a state-of-the-art platform in Stellare 2.0, which is fed by rich first-party data and AI automation to attract, assess, and approve our ideal customers. Our customers love us. They come back to us for future finance needs, which creates a great annuity stream for us, with minimal customer acquisition costs due to our existing direct relationship.
This allows us to effortlessly scale and serve more customers with our existing team. We then utilize our world-class funders and ABS programs, which ultimately leads to a 20% cash return on equity, of which we expect to be on that run rate during FY second half of FY 2025, which I'll go through in more detail later in the presentation. Now, turning to slide 6. To profitably build scale, it's clear that you need automation. For that, you need advanced automation and AI capability, and to do that successfully, you need more than just tech. It needs massive amounts of quality first-party data. First-party data means that the financial information is direct from the customer, not handpicked or pre-filtered by a broker. We consistently attract over 10,000 new customer accounts each month. All of this data is used to continually train our AI models.
This is a huge number of new customers for any business. This high volume of consumer financial data, combined with 10 years of historic data, effectively supercharges training our AI learning, so we can optimize for sophisticated, highly efficient marketing with platforms such as Google, giving us the right customers at low cost, and risk-adjusted income of 5%, gained through more accurate risk assessment of customers. This combination of data, AI, and automation built into our technology platform, has been a core feature of Harmoney since our inception. Now turning to slide 7. The power of Stellare's machine learning goes far beyond assessing customers at their point of loan application. It plays a significant role throughout the entire lifetime of our relationship with the customer. Harmoney is highly selective when it comes to audience targeting. We need to ensure our marketing spend is efficient and effective.
This is where sophisticated customer acquisition models play a key role, and again, where huge amounts of first-party consumer data is key. When these models are used with powerful digital marketing platforms like Google, Microsoft Ads, and Facebook, the result is highly targeted and cost-effective customer acquisition. What this means in practice is that we can acquire the right customers for the right price at scale, and critically, we can forecast this with a high level of reliability. Proof of our ability to target the right customer and the right product is reflected in our Google Reviews and Shopper Approved scores, with over 57,000 reviews at an average score of 4.8 out of 5. Moving to the second column, naturally, we work hard to ensure we deliver a great customer experience, so we can create annuity revenue as satisfied customers return with minimal customer acquisition cost.
This experience is underpinned by our automated, simple, and streamlined 100% online process. In the third column, our ability to scale remains an important factor in the Harmoney model, and our continued investment in technology remains a key enabler. Already in this year, we have achieved a cost-to-income ratio of 24%, which is exceptional. The diagram at the bottom of the page shows how all these factors combine to support the lifetime value of a customer. Our customer acquisition model helps us to attract the right customers. Our application and loan experience is highly tuned to customer satisfaction, so customers return to Harmoney for their future needs. Based on our New Zealand experience, this cycle is expected to lead to 140% in additional originations to the same customers at minimal customer acquisition cost.
For example, on average, every customer taking out an initial NZD 20,000 loan would return to Harmoney for an additional NZD 28,000. Now turning to slide 8, and then to slide 9 for our FY 2024 highlights. To highlight our key achievements for financial year 2024, this was our fifth consecutive half of Cash N PAT profitability and loan book growth, despite significantly higher interest rates. Our net interest margin for the final quarter of financial year 2024 was back over 10% for new business. We are now past the peak of the interest rate cycle, therefore, this higher margin will start to spread through the loan portfolio. Acquisition costs were down 14% year-on-year, and our automation drove continued efficiency gains, with our cost-to-income ratio down to a market-leading 24%.
We launched Stellare 2.0, which I'll talk about more on the following pages. In July 2024 , which was the first full month where all marketing channels were directed to Stellare 2.0, we experienced over 50% growth in loan originations compared to July 2023 . We successfully completed New Zealand's first-ever unsecured personal loan-backed ABS. A new warehouse was added, and three existing facilities were extended for a further two years, creating growth capacity of NZD 181 million. We refinanced our corporate debt and increased the limit by a further 50%, providing junior note capacity for up to NZD 200 million in additional loan book growth. Now turning to slide 10. Stellare 2.0 represents a significant leap forward in our technology from our previous peer-to-peer platform, to a new cloud-native banking solution used by globally recognized banks in Europe and the U.S..
Based on a modern cloud stack through our relationship with Google Cloud, it is more secure, with native horizontal scaling, continuous deployment, and high availability. Importantly, we've delivered a new core banking system together with our internally developed process automation across business and technology teams. This makes it easy for us to fully automate complex process flows across people and systems. We are launching Stellare 2.0, using a small fraction of its capabilities. However, it's designed to seamlessly accommodate a spectrum of financial products, ranging from deposits to home loans. Stellare 2.0 has been designed for enterprise performance, availability, and security at scale, while retaining the flexibility to help us deliver new innovative products and services in the future. Now turning to Slide 11.
We're excited to share that in July 2024, we moved 100% of our Australian new originations to Stellare 2.0. This is a great snapshot of the success story we're seeing in Stellare 2.0, as in July, where all traffic was directed to the new platform, we originated AUD 14.4 million of new business, representing a 50% uplift in volume compared to July 2023, when all traffic was directed to Stellare 1.0. This was achieved through advanced algorithms and rules with Stellare 2.0, to address segments of our customers that were underserved in the past. Previously, a customer applying for AUD 5,000 was assessed in a similar way to someone applying for AUD 70,000.
This meant that for the same 10,000 people, Stellare 1.0 would have only given an offer to 650 people per month. Under Stellare 2.0, this has increased to 1,300 people per month, which is a 100% increase. We offered nearly double as many loans under AUD 15,000 in July 2024 compared to July 2023. These are people who are good credits, would have likely been automatically declined in Stellare 1.0 because of that inflexibility. You may recall from our earlier presentations, that this was an area that we expected Stellare 2.0 to better serve our customers. Alongside these underlying improvements, the team has deployed a beautiful new application flow, which is driving better customer retention through the application process and helps us collect more accurate financial data. Now turning to Slide 12. These results for July are not a one-off.
With a platform and automation in place, we are now seeing consistent and repeatable results. This trend has been seen for twelve months now, while we have been progressively rolling out Stellare 2.0. But since July, we are at full throttle on Stellare 2.0, and this growth has continued into August. We've achieved automation and scalability safely by using 10 years of data to create the rules and data models, and the decisions are audited every day by credit officers. With Stellare 2.0, 90% of applications are completely automated. This has all been delivered, achieved by our in-house team, without any additional tech staff and costs. We've saved over 10% of costs, despite running Stellare 1.0 in New Zealand, alongside Stellare 2.0. And as we fully decommission our old platform, we expect to see further cost savings.
Improved conversion and origination volume hasn't come at cost of credit quality, with a 10% improvement in arrears and no change to the average Equifax scores, as a result of more personalized limits and risk rules. Alongside the automation, we also have a credit officer review every loan post-settlement to identify any potential improvements to rules for future decisioning in the platform. As we've covered on the previous slide, July was the first month where we had all new Australian borrowers assessed by Stellare 2.0. Conversion by dollars was up 50% on July 2023. This is showing the efficiency and reach of our new platform, as acquisition costs are down 30% on a cost per loan basis. Turning to Slide 13, I'll now hand over to our Chief Financial Officer, Simon Ward, who will take you through our financial results in more detail.
Thanks, David, and good morning, everybody. Please turn to Slide 14. I'll begin by going through each of these items briefly now, before going into more detail on the following slides. Starting with our loan book, which reached NZD 758 million, up 2% from last year, while revenue grew 15% to NZD 123 million, driven by both loan book growth and a higher average portfolio interest rate. Our net interest income percentage contracted 80 basis points to 8.8%, just short of our target 9%, driven by higher funding costs. However, with our fourth quarter new lending net interest margin above 10%, we're expecting our portfolio to return to the 9% during FY 2025.
Risk-adjusted income, which is our interest income after funding costs and incurred credit losses, contracted 120 basis points to 4.8%, driven predominantly by the reduced NIM just mentioned, and also slightly higher incurred credit loss rate, which I'll explain in more detail on a later slide. Offsetting this, acquisition costs were down NZD 1.7 million, 14% to NZD 10.6 million, as we moderated pursuit of new customer growth while consumers adjusted to higher costs and climbing market interest rates. Harmoney's high level of automation, which enable us to scale our loan book without having to similarly scale operating costs, continue to drive improvements in our exceptional cost-to-income ratio, which is now 24%, down from an already low 28% last year. Our statutory loss this year increased to NZD 13.2 million.
However, this includes a one-off NZD 9.5 million non-cash impairment expense on the retirement of our Stellare 1.0 platform, enabled by the successful launch of our new Stellare 2.0 platform. Statutory NPAT, normalized to exclude that one-off Stellare 1.0 impairment adjustment, was a loss of NZD 3.7 million, a 51% improvement on last year. The cost-to-income ratio of 24% has also been normalized to exclude that one-off impairment adjustment. However, it otherwise includes all operating expenses, including share-based payments, depreciation, and amortization. Finally, on this slide, Harmoney has generated a cash NPAT profit of NZD 0.7 million, which although down on last year, driven predominantly by the higher interest costs, represents our fifth consecutive half of both positive cash NPAT and continued loan book growth.
This represents stability and consistency, which we've been pleased to demonstrate through this tightening phase of the interest rate cycle, and which we believe puts us well placed to capitalize on Stellare 2.0 in the next phase of the cycle. Turning now to Slide 15, to look at our loan book and revenue in a bit more detail. Higher market interest rates through the year have influenced many consumers to defer activities that require finance, and where they haven't, in some cases, higher interest rates have lowered their borrowing capacity. These factors have constrained book growth this year to 2%. However, we believe that as consumers adjust to higher interest rates and as rates start to ease, this will unlock some pent-up demand.
While book growth has been moderate, revenue has grown 15% on last year to NZD 123 million, driven by a combination of book growth and a higher average portfolio interest rate, as interest rate increases passed through on new lending have grown to a higher proportion of the portfolio. The average interest rate this year was 16.2%, up from 15.5% last year. Our average new lending rate through our most recent quarter has been 18%. This, together with the continuing amortization of lower rate earlier year loans, should continue to increase our average portfolio rate for some time. Now turning to Slide 16, providing detail on our funding. This year, Harmoney renewed and added to its bank warehouse facilities, provided by three of the Big Four Australian banks.
We also issued New Zealand's first public asset-backed securitization of unsecured personal loans, as well as refinancing and upsizing our corporate debt facility. Our average funding rate increased a hundred and seventy basis points to 7.7% this year, driven predominantly by funding market increases, but also partially driven by achieving higher leverage within some of our facilities on the proven quality of our loan portfolio. Harmoney's balance sheet is very capital efficient, with borrowings equivalent 95% of the loan book. With unused warehouse capacity of NZD 181 million, undrawn debt capacity of NZD 7.5 million, and over NZD 20 million of unrestricted cash on hand, Harmoney is well capitalized for loan book expansion in the coming year. Gearing coverage ended this year at 85%... Turning to slide 17, focusing on credit performance.
The deep consumer data provided by our consumer direct model powers our AI credit models. This has enabled us to build a prime loan book of resilient borrowers, with 74% employed in either professional, office, or trade roles, and 87% aged over thirty. Further demographic detail on the loan book is provided in the appendix to this presentation. Credit losses in the second half of the year have come down to 4% from 4.2% in the first half of the year. As discussed in our first half results, an earlier, less accurate Australian scorecard was transitioning through its peak hazard for loss period during that first half, which elevated losses in that first half. As the two charts on the top right show, the rest of the portfolio now forms the much larger component and has a much lower static loss rate.
As a consequence, and as expected, losses have begun trending back down in the second half, reaching that 4% back within our target 3%-4% range. Our Risk-adjusted income percentage, which is our income margin after both funding costs and credit losses, has reduced to 4.8% this year, down from 6% last year, in part driven by this year's uptick in the incurred credit loss rate, but to a greater extent, driven by the increase in the average funding rate discussed on my last slide. However, this margin has already returned above 6% for new business written in the fourth quarter, with higher interest rates passed through on new lending and as funding costs have begun to ease.
The chart on the bottom right shows our 90-plus day arrears have also returned towards historic lows, finishing the year at 43 basis points, around a quarter of the current Australian personal loan market average. Now turning to slide 18, providing more detail on our acquisition costs. Harmoney's algorithms have continued working in partnership with Google's to cost effectively seek out those customers most likely to take a Harmoney loan. This consumer direct approach provides us with incredible flexibility and agility to adjust and fine-tune our acquisition costs in response to market conditions. This year, while continuing to grow our loan book, we've managed to reduce acquisition costs by 14% to NZD 1.7 million. The chart on the right, which is one we show regularly, shows both our total acquisition spend and our ratio of cost per dollar originated.
You'll see that the ratio has steadily trended down over the years with a slight uptick this year. One of the major benefits of Harmoney's consumer direct model is a key driver of both the overall downward trend in that ratio and the slight uptick this year. That being, that due to the great experience we provide, our customers tend to return to us the next time they want a loan, and because of our existing direct relationship with them, those subsequent loans have near zero additional acquisition costs. This year, these existing customer originations have also been impacted by higher market interest rates. However, lower originations to these customers are not accompanied by a reduction to the acquisition costs, as there are next to none, and this has driven the small uptick shown in the chart.
This coming year, off the back of both increased new and existing customer originations, driven by both increased conversion from Stellare 2.0, as well as improving market sentiment, we're confident that the historical downward trend of this marketing efficiency ratio will resume. Now turning to Slide 19, looking at our operating expenses. A key feature of Harmoney's Stellare platform has always been its high levels of automation, which has been driven, which is even further enhanced this year with the successful launch of Stellare 2.0. This automation again enabled Harmoney to scale its loan book much faster than its operating expenses, with this year's operating expenses actually reducing 4% while the loan book continued to grow.
This is shown in the chart on the right, with the loan book growing from NZD 744 million last year to NZD 758 million this year, while the cost-to-income ratio has fallen from an already low 28% last year to 24% this year. We fully expect this scale effect to continue in future periods as we continue to grow our loan book. This ability of our Stellare platform to scale has enabled Harmoney to deliver its fifth consecutive half of both loan book growth and Cash NPAT profitability, even during this tightening phase of the interest rate cycle. This is what underpins our confidence in achieving our 20% Cash ROE run-rate target in the second half of 2025.
With that, turning to Slide 20, I'll hand you back to David to take you through our outlook.
Thanks, Simon. Continuing now to our outlook, please turn to slide 20 and then slide 21. Harmoney is well positioned to capitalize on the capabilities of our new platform and into its next phase of growth. We have four main strategic priorities for financial year 2025: the core platform, continued improvement of the conversion of customers, developing new opportunities, and increasing margin. Within the platform, our biggest immediate focus will be bringing the power of Stellare 2.0 to our New Zealand customers. Importantly, we've always built the new platform with multiple countries in mind, so this is not anticipated to be a huge piece of work. However, as we did in Australia, we will continue to conservatively roll out the platform and make sure that the system is working well before we roll out 100% of the volume....
Following this, we'll be able to fully retire Stellare 1.0, and we expect to drive further costs out of the business. The core of the platform is in a great state, so the further developments here will be taking things to the next level, including next-generation credit scenario back-testing, to allow us to easily test new processes and credit models against historical applications. On conversion, the game changer has been the real-time visibility into every interaction our customers have with Harmoney and the data feeding back into our new product development. This has allowed us to release over 1,200 times in the last year, a mixture of small and big updates to increase conversion, and we'll continue that effort in the coming year.
However, there is also huge scope for increased use of AI tools to personalize the application experience and offers that customers receive to help more customers. Having the platform live also presents us with a number of new capabilities that we can take advantage of. We already collect most of the data we need to assess the customer for multiple products, so the next step is to expand the lending products we have available. While it's too early to get into the specifics at this stage, we have some exciting new partnership opportunities that we're working through that will leverage the flexibility and capabilities of Stellare 2.0.
The personalization that new platform affords us not only allows us to ensure more people get to a great offer when they come to us, like, it also means that we can achieve our target net interest margin of 10% through a new pricing optimization engine that we're working on, as well as increasing sophisticated process automation that delivers a better customer experience at scale and with lower costs for us. Now, turning to slide 22. In terms of our overall outlook, we have split this into three areas, being interest rates and asset quality, growth outlook, and specifically first half 2025 and second half 2025 financial years.
With respect to interest rates and asset quality, a reminder, Harmoney has the ability to pass through targeted interest rate increases and decreases through our sophisticated risk-based pricing model to achieve our 10% net interest margin. Funding costs are managed with a diversified funding panel of three of the big four banks and mezzanine funders, together with an ABS program in Australia and New Zealand, while 85% of borrowings are hedged. It's a high quality, diversified loan book, of which 74% are employed in either professional, office or trade roles, together with an extremely low 43 basis points at 90 day arrears levels. With respect to the growth outlook, we expect growth to remain strong due to the large total addressable market. Harmoney's consumer direct model is taking market share from banks, with plenty of room to grow in the AUD 150 billion dollar market.
Harmoney has been and continues to work with Google to implement further AI technology to enhance customer experience, lowering customer acquisition costs, and to further reduce our cost-to-income ratio. The rollout of Stellare 2.0 across the entire business is set to increase revenue, lower costs, and drive higher profitability. We strategically invested in this project through a challenging, higher interest rate environment and cost pressures on consumers, and have carefully managed costs to make sure we kept growing conservatively through that period. So we're excited for the coming years, as we're well- very well positioned to benefit from that investment now. Specifically, in terms of the financial year 2025 outlook, it appears that the interest rate cycle has reached its peak in both markets, and Harmoney has now proven that its business model is highly resilient through all phases of the interest rate cycle.
Therefore, we expect to see in first half 2025 completion of Stellare 2.0 rolled out in both countries to set us up for significant growth in second half 2025 and beyond. We expect to see significant acceleration in loan book growth, net interest margin on the loan book to return to 9%. You will note from our earlier comments that the fourth quarter 2024 was over 10% net interest margin on new business, which will spread through the loan book over time. We also expect to see Cash NPAT growth in financial year 2025, and also a 20% cash return on equity run rate achieved in second half 2025. Now, turning to slide 23. As I just said, a key focus for us is targeting a 20% cash return on equity run rate in second half 2025.
So I wanted to try and show you this as simply as possible, and why we have conviction in achieving this target. The diagram on the right shows a simple illustrative scenario of us getting to this goal. In financial year 2023, we achieved a risk-adjusted income margin of 6%. Risk-adjusted income is our income after funding costs and credit losses, divided by our average loan book. This year, that margin reduced to 4.8%, driven predominantly by higher market funding costs discussed earlier by Simon. However, as also mentioned, our margin is already widening as higher interest rates pass through on new lending grow as a proportion of the loan book, as funding costs begin to ease, as credit losses, and as credit loss rates improve. This is illustrated by the risk-adjusted margin on fourth quarter 2024 new business already being above 6%.
While it will take some time for this to reprice through the entire loan book, it will lift the portfolio risk-adjusted margin through financial year 2025. The scenario on the right adopts a midpoint of the financial year 2023 and 2024 risk-adjusted margins at 5.4%. At that margin, Harmoney could achieve its 20% cash return on equity target from an average loan book of NZD 819 million, which is only 8% higher than the 30 June 2024 balance of NZD 758 million. For added context, the two-year cumulative annual growth rate for Harmoney's loan book is 14%.
Achieving that 20% cash return on equity assumes a 5% inflationary growth rate on our financial year 2024 customer acquisition and cash operating expenses. We're very comfortable with this, as historically, these costs have been stable or falling, even in a high inflation environment, due to efficiency gains, including from high levels of automation in Stellare platform. The scenario shown is just one example of a potential pathway to our target of a 20% cash return on equity run rate in the second half 2025. Clearly, we could also reach that target with a different mix of risk-adjusted income margin or loan book growth, but I'm hopeful that by setting out this simple scenario based on historically achieved margin, growth, and cost inputs, you can see how credible and within reach we believe this target to be. That concludes the presentation for today.
Now turning to slide 24. I'd just like to remind you that we recently set up our new Investor Hub, which is available at harmoney.com.au/invest. We've created this to provide shareholders and potential investors with a centralized place for all Harmoney investment resources. You'll find updates, announcements, reports, videos, and more, plus shareholders can ask questions or share comments directly with the Harmoney team. Please join our Investor Hub. We will now turn to answering your questions. Just a reminder that you can submit a question on the webcast by typing into the "Ask a Question" box and then hitting "Submit.
Thank you. Your first phone question comes from Ian Munro with Ord Minnett.
Good morning, Dave. Good morning, Simon. Thanks for taking my questions. Might just ask three, if that's all right, fellas. Just firstly, just on your comments around the funding, margins on the warehouse facilities, can you just perhaps give us a sense of the trend in the market, whether you've got any facilities up for renegotiation over the next twelve months? Secondly, loan book size, just in terms your intention to grow that into FY 2025, you know, are we fair to assume a more aggressive rate of growth now that Stellare 2.0 is up and running?
And then thirdly, just on the partnerships you alluded to, Dave, is there anything you could perhaps flesh out a little bit more as to, you know, what are your options for those partnerships and what that might mean for the business? Thank you.
Thanks, Ian. In terms of funding availability, as I said in the presentation, we just renewed set a new facility up in the year, and we renewed three of the warehouses for another two years. So there's actually we might have one due at the end of June next year. So we're, you know, we're fairly got a fair way, which is good, you know, with very big support from three of the Big Four banks. Plus, we obviously run our securitization programs to, you know, empty some of those warehouses out along the way. If you follow the market at all, the ABS market in this space has been really, really strong. There's been new issuers and doing it at really, really cheap rates, actually, really good margins for the issuers.
Certainly plenty of funding capacity and funding availability. Loan book growth, yeah, look, we expect the loan book growth to really start moving now, particularly in the second half, as we bring everything onto Stellare 2.0. And, you know, we think that's, you know, that's where our... Yeah, that's what we're here for, right? We're here to grow loan book at good margin. And, yeah, that's what the key focus of the business is and what all our – what, you know, what, you know, I think our incentives are based around and, what, you know, shareholder value is based around.
So, you know, it's the reason I'm very excited about, sort of, top of the call was, you know, this is something we've been working on for eighteen months, just over eighteen months, and we've finally got it. It's like having a, you know, a Ferrari in the garage when we used to have a, you know, I don't know, Falcon. And lastly, on the partnerships, yeah, look, you know, they've reasonably progressed, probably can't share too much detail, but they're working with some partners that, you know, we think are pretty exciting. Hopefully have an update on that, you know, maybe the quarter, but certainly in the next half.
Thanks, Dave.
Cheers, Ian.
Okay, got some questions online. Can you give some examples of how you stress Stellare 2.0's improved application approval rate to ensure that it's not overly optimistic in approving applications compared to Stellare 1.0?
So first of all, and I mentioned in the presentation, we do a lot of back testing against any changes that we put through, and we do that against not only approved deals, we do that against declined deals as well. That gives us a great understanding of at what change that we make, what impact that could have on a customer we wrote in the past or a customer that we didn't write in the past. That's critical when you're running a financial services business like ours.
Probably the key point, as I said, we haven't changed credit rules per se. It's just that the system now targets uses more flexibility on smaller loans. It doesn't sort of assess them for. If you go back to our peer-to-peer days, we were very incentivized to you know, lend as much as possible and that was you know, that was the way the model worked, providing it was done responsibly. Whereas the smaller deals weren't really as a key focus for us, and the system was all sort of built around that model. Well, we're still gonna be writing the same deals that we do today at the higher level.
It just allows us to get into, you know, a market where our system was probably cutting people off that we shouldn't have. We still apply minimum Equifax scores across all our loans. There's been absolutely no change in that. We use an income to loan percentage, and we use various other risk factors through the way there. Just because it's a smaller loan amount, you know, remember, our average loan size was NZD 20,000. That will probably come down a little bit over time as we have more of a weighting to the smaller loans. But we're not, you know, we still write the same and more actually, because of all the other things we mentioned about how we've got much better insight into the customers.
So we're writing a lot more loans over NZD 20,000 as well. And it’s really. Everyone, you know, will talk about their application and that, and I'm not sitting here saying our application is better than others, but certainly as far as what the consumer sees. But as far as what we see and what we're able to work understand where the customer's at and speak to them in the process, I think that's really advanced, and that is helping customers get through a process where they may have got stuck in the past, not for credit reasons, but for operational reasons.
Okay.
So I'm just reading through the questions.
Do you feel you might need a modest equity capital raising if growth goals are achieved in 2025?
The short answer is no. We have sufficient cash in this business. We've got NZD 21 million at the full year. We've got NZD 7.5 million of undrawn corporate debt. You know, it would have to be, and the business is profitable, so it's actually spinning off cash. So, you know, that would be in a scenario of, you know, super growth, like, you know, talking, you know, probably 100%, you know, that sort of level. It's not something that we're, you know, it's certainly not something we have in our forecasts and something that I don't plan on doing, unless there was some highly strategic reason.
Okay. On slide 12, can you please elaborate on the 90% automated decision? Do you mean... Sorry, it's not moving.
Do you mean 90% of the application approvals are fully automated from online applications to loan approval without any credit officer's manual assessment review?
Isn't this too risky for arrears and losses? Not really, to be honest. As I said, using humans is actually a lot riskier than using systems. If you've built your system properly, you get consistency of decisioning. Humans, you know, everyone's different, and everyone has different concerns come into the office with, you know, whilst they might use the same credit policy, it can actually change between people, quite a bit. So we've deliberately built the processes in.
We actually have credit officers, as I said, review every deal post-settlement to make sure that the loans that the system is approving is consistent with what they would approve and also what the system has been set up to do. And that's a key feature, and that's probably why we took the twelve months in rolling it out. You'll recall, we actually wrote our first deal about a year ago in Stellare 2.0. But we've taken a year, basically making sure those rules are operating correctly without putting major volume through the platform. So, yeah, look, it's the way the model works, it's how we get the automation, and you can see the performance, you know, the Equifax scores are the same. The arrears are actually lower, and, you know, we, we've got some...
I'd actually say we've probably got some tighter rules in there, you know, that, you know, around some things that we've learned throughout the way. And I prefer having a system make decisions, providing it's been architected properly and built properly and the appropriate risk rules, than having a lot of different humans reviewing deals.
Next question: Do you really mean that the 20% will occur by June 2025 to take effect from first half 2026?
No, we will be on the 20%. Our stated target is to be on 20% cash return on equity during second half 2025. So it won't be for the financial year 2025. It won't necessarily be for the whole second half of 2025, but we'll certainly be on that run rate in there.
I guess you could imply from that question that we will be on that from first half 2026. You know, that's right . You know, assuming it keeps going the way we would expect it to, that's right in that regard. But you'll, that's, we're pretty confident about that.
How do you achieve that improvement in NIM? Do you get benefits on margins from funders or customers?
I think as Simon sort of said, you know, we have increased our customer rates a little bit. Remember, we don't borrow off the thirty-day rate; we borrow off the swap rates, so they have started to trend down. The rates are actually inverted at the moment. You know, the longer term, we borrow off two to three-year rates.
So, you know, we're starting to see that margin come through the book, but obviously we do have 85% of the book hedged. So, you know, we've got older swaps that are locked in against pricing, so it does take some time to flow through it. But as you can see from our fourth quarter, we are already over 10% NIM, and, you know, we'll have, you know. That's always been our target. And, you know, as that newer business starts to become a bigger proportion of the loan book over the year and over future years, you know, that makes the, you know, that helps us with that target.
Okay. What is your reliance on Google for customer acquisition, and are these per client win costs?
I'm sure that replies to the first part of the question. Yeah, we do use. A lot of our originations come through Google and have been since the, you know, for seven, eight years now. You know, it's the biggest. It's got 95% market share when it comes to search. So we do a lot with Google. They're certainly our biggest channel for origination. But we, there are other channels we do as well in a direct nature. But yeah, so we absolutely continue to work with Google, and I think we're certainly one of the, well, probably the largest one that solely works with Google in Australia and New Zealand, in this space. Do you wanna do the New Zealand cash impact one, Simon?
Yeah. So, New Zealand cash impact was negative this financial year. Assuming interest costs were slow to change, how much would implementing Stellare improve the result? So I guess the first thing I'd point out is that, the New Zealand cash impact is shown as negative, because the bulk of our operating costs are actually based in New Zealand, is where our offices are and our tech development team is based. So it's very much a feature of Harmoney's business that our cost base is in New Zealand, and yet we are able, with very little additional cost, able to grow the Australian market on top of that.
So we very much view the business overall as a group, but when you break it down in that segment note, the main driver of that difference between Australia and New Zealand cash impact is really just 'cause this is where the bulk of the costs, the operating costs are. We definitely do expect Stellare 2.0 to uplift our originations in New Zealand. The level of increase in percentage terms may not be quite as big as Australia, because we are already much more established in the New Zealand market, and it is a smaller addressable market than Australia. But nevertheless, we do expect to see significant conversion improvements when we bring that model to bear in New Zealand.
Great. Thanks, Simon. The next question: Could there be a further write-down once Stellare 2.0 is rolled out in New Zealand in first half 2025, or it all been written off completely?
Yeah, good question. It's all been written off completely. The only cost we're carrying now is about NZD 4 million or so, and that's all in relation to Stellare 2.0, so there'll be no further write-off. Obviously, we made that decision this half based on the fact that we had full confidence that we were gonna run Stellare 2.0 across the whole business and was operating, so it made sense to make that write-down.
Stellare 1.0 could have continued on if we wanted it to, but hopefully you can see already that the performance and the things that we can now do with Stellare 2.0 are, you know, they're light years apart from where we were. Not that Stellare 1.0 was great, you know, it got us to 10 years. It's our 10-year anniversary of the company next month. So, you know, it got us all there. But like anything, you know, we are a tech business, and platforms need to be replaced with newer technology.
Okay, next question: Congrats on the progress on costs. Can you give some thoughts on where you think the cost-to-income ratio could look like a couple of years out?
Look, I think...
I'm not gonna give specific guidance on that, but we have a, you know, we're not adding costs to this business outside of, you know, cost to collect as the book gets bigger, and, you know, obviously, that's fairly small cost in the scheme of things. This business is a platform, and we've said that right from the start, so we don't have any vision of increasing cost in the business. Simple as that. We run a pretty tight ship here, and I think we, you know, we've proven that over years, so we'd expect that over, you know, years to come, keep coming down. What are the expected impacts of transitioning existing Stellare 1.0 book to Stellare 2.0 to be in FY 2025?
I think it's all included in our outlook. We're not. Like, there obviously is some internal time that's spent on doing that. Obviously we've built it with this in mind, that there will be some sort of you know, existing customers moving over. But given our loan book is a fully amortizing product, it actually, the transition is actually not that complicated, compared to, say, if you had around like a credit card book or something like that, where you had, you know, millions and millions of transactions, all with different terms attached to it and the like. It's fairly simple to you know, move a book that you've got a clear pay down period over to a new platform.
But that's something that will be going in the background, and I don't expect that to really be seen by anyone, and customers won't even know. It'll just be moved in the background, which I think is the way you want to do these things. Has Harmoney considered expanding into cash managing apps and early wage or tax access products? Not as such. Not really. Like, I understand there's some value around, you know, employer type channels and things like that, but no, we're not. That's not in the sort of... Certainly not in the roadmap.
I think that's getting towards the end of them. Just a question. Have that one. Cover those first two. Cover those three.
Sorry, guys, just got lots of questions on here, so I've got to do this in real time. We're not. It's not recorded, it's live, obviously.
You have less than 1% of the 150 billion dollar market. What would be your real aspirational target in market share to be in three years' time? Look, we don't. We obviously have, you know, some pretty decent targets now, particularly off the back of having the platform. And I know we've talked about this endlessly, but you have to remember that Harmoney is a platform business. We have direct relations with customers. We don't have sales teams where people go out and bring business to us.
So our platform is particularly more important than, say, another business model that the platform kind of just facilitates the transaction. Our platform is everything. So with that, we're now able to do. You know, as I said in the presentation, we've got a state-of-the-art core banking platform. We've got all our process automation on top of that, which now gives us a huge potential to go into that market, and we will continue to get as much of that as we can. And, you know, whilst I might have internal targets for the team, we just see it as we don't need to go outside our core. We know that if it's just between Australia and New Zealand and in the personal lending space, there's huge growth.
And I believe that's much lower risk to any organization than sort of having a strategy where you're going to other countries or you're going into, you know, let's just say, commercial lending, for example, where it's two steps away from your core. That's not something we need to do. We're hitting start. We've got a huge market right in front of us, which is, you know, part of it, part of what we're good at and where our core lies .
If you decided to stop growing the loan book, how much cost could be saved, both OpEx and provision for growth? How much cash would the business spin off?
Great question, and I think, you know, anyone that's good with Excel could probably figure that out.
It's, you know, you know, it does spin off a lot of money because you've got, you know, 750-odd million at a 9% or 10% margin on that. In that scenario, certainly not a scenario that we're planning on or doing, but it would be significant. I'm not gonna quote the number, but you could work that out in a model if you were to do that, and I think it would be quite high.
Once Stellare 2.0 is rolled out in New Zealand, how far away are the new products using the new capability?
Look, we are. Part of the team is working. The product team does work on innovation, and that's continually going on at the moment.
But we'd like to have some similar, some products out in the next year. But, you know, the focus really is making sure we roll New Zealand out properly, and we also get continuing to increase that conversion in Australia. The fact that we're, you know, we're now offering 1,300 customers an offer as opposed to 650, that's huge. You know, and that, and that's something we want to. You know, why can't that be 1,500? Why can't that be 2,000? We keep getting 10,000 customers a month, you know, it, it's a huge opportunity, and that's where I think Harmoney is differentiated. We don't need to, you know, go and incentivize intermediaries or, or brokers or, or the like, to send us more business.
We've actually already got the customers coming to us. So, you know, with the new platform, we're able to get to those a lot quicker and a lot better. And we have a lot. And that's why the offers are up. You know, they're getting a better experience. We're seeing better insights to where they're at. If they, you know, make a mistake in the application, which is obvious, you know, they said they earn NZD 5,000 a month, and it should be a week, you know, like things like that, that wouldn't, might not have been seen before and declines might have been made. Now we're seeing that in real time, and this is the difference between using a 2024 technology and a 2014 technology.
We're lucky, we don't have the legacy of, you know, a lot of financial businesses that have to employ massive teams to do that. We've done all this with our in-house team. CapEx is pretty much the same if you look year on year, and you know, we don't expect that to go up either. That's just it's our run rate of the business, and we have been at that level for some time. I think that's it for the questions, so we might wrap up the call there.
That does conclude our conference for today. Thank you for participating. You may now disconnect.